Article(id=1327018072738738527, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327017934683222337, articleNumber=null, orderNo=null, doi=10.7544/issn1000-1239.202220749, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1681056000000, receivedDateStr=2023-04-10, revisedDate=1744214400000, revisedDateStr=2025-04-10, acceptedDate=null, acceptedDateStr=null, onlineDate=1794315363344, onlineDateStr=2026-11-10, pubDate=1759248000000, pubDateStr=2025-10-01, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1794315363344, onlineIssueDateStr=2026-11-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1794315363344, creator=13701087609, updateTime=1794315363344, updator=13701087609, issue=Issue{id=1327017934683222337, tenantId=1146029695717560320, journalId=1324759914146529478, year='2025', volume='62', issue='10', pageStart='2397', pageEnd='2650', issueExtLink='null', onlineDate='null', pubDate='1759248000000', pubDateStr='2025-10-01', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1794315330429, creator='13701087609', updateTime=1794376951272, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1327276391986475384, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327017934683222337, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1327276391986475385, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327017934683222337, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=2583, endPage=2594, ext={EN=ArticleExt(id=1327018072919093600, articleId=1327018072738738527, tenantId=1146029695717560320, journalId=1324759914146529478, language=EN, title=A Graph-Based Interpolation Sequential Recommender with Deformable Convolutional Network, columnId=1325489891789910943, journalTitle=Journal of Computer Research and Development, columnName=Artificial Intelligence, runingTitle=null, highlight=null, articleAbstract=

Sequential recommendation systems aim to predict users’ next actions based on the preferences learned from their historical behaviors. There are still fundamental challenges for sequential recommender. First, with the popularization of online services, recommender need serve both the warm-start and cold-start users. However, the most existing models depending on user-item interactions lose merits due to the difficulty of learning sequential dependencies with limited interactions. Second, users’ behaviors in their historical sequences are often implicit and complex due to the objective variability of reality and the subjective randomness of users’ intentions. It is difficult to capture the dynamic transition patterns from these user-item interactions. In this work, we propose a graph-based interpolation sequential recommender with deformable convolutional network (GISDCN). For cold-start users, we reconstruct item sequences into a graph to infer users’ possible preferences. To capture the complex sequential dependencies, we employ the deformable convolutional network to generate more robust and flexible filters. Finally, we conduct comprehensive experiments and verify the effectiveness of our model. The experimental results demonstrate that GISDCN outperforms most of the state-of-the-art models at cold-start conditions.

, authors=Xinyue Liu1, Hailian Yin4, Yalei Zang1, Wenlong Wu1, Junnan Zhuo1, Fengru Xu1, lüying Chen1, Weihua Ma5, Bohan Li1, 2, 3, authorsList=Xinyue Liu, Hailian Yin, Yalei Zang, Wenlong Wu, Junnan Zhuo, Fengru Xu, lüying Chen, Weihua Ma, Bohan Li, authorCompany=null, correspAuthors=null, authorNote=

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

born in 1979. PhD, associate professor. Her main research interests include multidisciplinary design optimization and machine learning

born in 1996. Master candidate. Student member of CCF. His main research interests include sequential recommendation and graph neural networks

born in 2001. Master candidate. Student member of CCF. His main research interests include large language model and recommendation systems

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

born in 2001. Master candidate. Student member of CCF. Her main research interests include recommendation systems and large language models

born in 1998. Master candidate. Student member of CCF. Her main research interests include cross-domain recommendation and sequential recommendation

born in 1960. PhD, professor. His research interests include multi-computer systems, embedded systems and applications, and computer control and intelligent systems

born in 1979. PhD, associate professor. Distinguished member of CCF, member of ACM. His main research interests include spatio-temporal database, knowledge graph, and recommendation systems

, correspAuthorsNote=null, copyrightStatement=Copyright © 2025 Journal of Computer Research and Development. All rights reserved., copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=null, magXml=null, pdfUrl=null, pdf=null, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=null, mapNumber=null, fund=null), CN=ArticleExt(id=1327018077331501442, articleId=1327018072738738527, tenantId=1146029695717560320, journalId=1324759914146529478, language=CN, title=基于图插值和可变形卷积网络的序列推荐, columnId=1325489892104483745, journalTitle=计算机研究与发展, columnName=人工智能, runingTitle=null, highlight=null, articleAbstract=

序列推荐系统(sequential recommendation system,SRS)旨在基于用户的历史行为偏好预测下一步行为. 尽管针对序列推荐提出了许多有效的方法,但仍然存在根本性的挑战. 首先,随着在线服务的普及,推荐系统需要同时服务于热启动用户和冷启动用户. 然而,由于难以从交互数据有限的序列中学习到有效的序列依赖关系,大多数依赖于用户-项目交互的现有模型失去了优势. 其次,由于现实中用户意图的可变性和主观随机性,用户在其历史序列中的行为往往是隐含和复杂的,很难从这些用户-项目交互数据中捕获这种动态转变模式. 提出了一种基于图插值和可变形卷积网络的序列推荐(graph-based interpolation sequential recommender with deformable convolutional network,GISDCN)模型. 对于冷启动用户,将序列对象重新构建成图,并提取全局序列中的知识来推断用户可能的偏好. 为了捕捉复杂的顺序依赖关系,使用可变形卷积网络来生成更健壮和灵活的卷积核. 最后,在4个数据集上进行了综合实验,验证了模型的有效性. 实验结果表明,GISDCN优于大多数主流的模型.

, authors=刘昕悦1, 尹海莲4, 臧亚磊1, 吴文隆1, 卓俊男1, 徐凤如1, 陈吕莹1, 马维华5, 李博涵1, 2, 3, authorsList=刘昕悦, 尹海莲, 臧亚磊, 吴文隆, 卓俊男, 徐凤如, 陈吕莹, 马维华, 李博涵, authorCompany=null, correspAuthors=null, authorNote=

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

1979年生. 博士,副研究员. 主要研究方向为多学科设计优化、机器学习

1996年生. 硕士研究生. CCF学生会员. 主要研究方向为序列推荐、 图神经网络

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为大语言模型、推荐系统

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、大语言模型

1998年生. 硕士研究生. CCF学生会员. 主要研究方向为跨域推荐、序列推荐

1960年生. 博士,教授. 主要研究方向为多机系统、嵌入式系统及应用、 计算机控制与智能系统

1979年生. 博士,副教授. CCF杰出会员,ACM会员. 主要研究方向为时空数据库、知识图谱、推荐系统

, correspAuthorsNote=null, copyrightStatement=版权所有 © 2025 计算机研究与发展编辑部, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=Jq/2RcVFpC5viIoE3hGvZw==, magXml=2KffPx0iywqwa5Nq1hS2/Q==, pdfUrl=null, pdf=yFBjmeFvzgPP0bMdbcJsfA==, pdfFileSize=3026199, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=TbLKWScdUjPQE/mwm/niJw==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=XpnfHmyU3ghIiBtXS5CmNg==, mapNumber=null, fund=null)}, authors=[Author(id=1327019126175936770, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126247239940, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126175936770, language=EN, stringName=Xinyue Liu, firstName=Xinyue, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

"}, bioImg=null, bioContent=

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019126326931717, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126175936770, language=CN, stringName=刘昕悦, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"hv9eUo+PLu0dbaQtB71RRA==","content":"

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

"}, bioImg=hv9eUo+PLu0dbaQtB71RRA==, bioContent=

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019126402429191, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126473732361, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126402429191, language=EN, stringName=Hailian Yin, firstName=Hailian, middleName=null, lastName=Yin, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=4, address=4Key Laboratory of Advanced Technology for Small and Medium-sized UAVs (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology, Nanjing 210006, bio={"content":"

born in 1979. PhD, associate professor. Her main research interests include multidisciplinary design optimization and machine learning

"}, bioImg=null, bioContent=

born in 1979. PhD, associate professor. Her main research interests include multidisciplinary design optimization and machine learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019126540841226, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126402429191, language=CN, stringName=尹海莲, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=4, address=4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006, bio={"img":"uUl4gZx//Y05SdHXvI4txw==","content":"

1979年生. 博士,副研究员. 主要研究方向为多学科设计优化、机器学习

"}, bioImg=uUl4gZx//Y05SdHXvI4txw==, bioContent=

1979年生. 博士,副研究员. 主要研究方向为多学科设计优化、机器学习

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019126020747515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=4, ext=[AuthorCompanyExt(id=1327019126029136124, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126020747515, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4Key Laboratory of Advanced Technology for Small and Medium-sized UAVs (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology, Nanjing 210006), AuthorCompanyExt(id=1327019126037524733, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126020747515, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006)])]), Author(id=1327019126612144396, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126687641870, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126612144396, language=EN, stringName=Yalei Zang, firstName=Yalei, middleName=null, lastName=Zang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 1996. Master candidate. Student member of CCF. His main research interests include sequential recommendation and graph neural networks

"}, bioImg=null, bioContent=

born in 1996. Master candidate. Student member of CCF. His main research interests include sequential recommendation and graph neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019126771527951, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126612144396, language=CN, stringName=臧亚磊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"4WlZ15tnNpA0m2lAtF6obg==","content":"

1996年生. 硕士研究生. CCF学生会员. 主要研究方向为序列推荐、 图神经网络

"}, bioImg=4WlZ15tnNpA0m2lAtF6obg==, bioContent=

1996年生. 硕士研究生. CCF学生会员. 主要研究方向为序列推荐、 图神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019126834442513, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126905745683, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126834442513, language=EN, stringName=Wenlong Wu, firstName=Wenlong, middleName=null, lastName=Wu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2001. Master candidate. Student member of CCF. His main research interests include large language model and recommendation systems

"}, bioImg=null, bioContent=

born in 2001. Master candidate. Student member of CCF. His main research interests include large language model and recommendation systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019128591855893, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126834442513, language=CN, stringName=吴文隆, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"8WK03dFG1iDz2SBDeToL0Q==","content":"

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为大语言模型、推荐系统

"}, bioImg=8WK03dFG1iDz2SBDeToL0Q==, bioContent=

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为大语言模型、推荐系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019128709296407, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019128780599577, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128709296407, language=EN, stringName=Junnan Zhuo, firstName=Junnan, middleName=null, lastName=Zhuo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

"}, bioImg=null, bioContent=

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019128843514138, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128709296407, language=CN, stringName=卓俊男, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"dFoSOgoJN8GJlunF1GVtMw==","content":"

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

"}, bioImg=dFoSOgoJN8GJlunF1GVtMw==, bioContent=

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019128902234396, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019128981926174, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128902234396, language=EN, stringName=Fengru Xu, firstName=Fengru, middleName=null, lastName=Xu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2001. Master candidate. Student member of CCF. Her main research interests include recommendation systems and large language models

"}, bioImg=null, bioContent=

born in 2001. Master candidate. Student member of CCF. Her main research interests include recommendation systems and large language models

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129044840735, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128902234396, language=CN, stringName=徐凤如, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"8v45a+G3QHfVWIWmrLuINg==","content":"

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、大语言模型

"}, bioImg=8v45a+G3QHfVWIWmrLuINg==, bioContent=

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、大语言模型

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019129120338209, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019129191641379, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129120338209, language=EN, stringName=lüying Chen, firstName=lüying, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 1998. Master candidate. Student member of CCF. Her main research interests include cross-domain recommendation and sequential recommendation

"}, bioImg=null, bioContent=

born in 1998. Master candidate. Student member of CCF. Her main research interests include cross-domain recommendation and sequential recommendation

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129250361636, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129120338209, language=CN, stringName=陈吕莹, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"HQogqnbpsi5m6luJlK6dtA==","content":"

1998年生. 硕士研究生. CCF学生会员. 主要研究方向为跨域推荐、序列推荐

"}, bioImg=HQogqnbpsi5m6luJlK6dtA==, bioContent=

1998年生. 硕士研究生. CCF学生会员. 主要研究方向为跨域推荐、序列推荐

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019129317470502, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=7, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019129376190760, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129317470502, language=EN, stringName=Weihua Ma, firstName=Weihua, middleName=null, lastName=Ma, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=5, address=5College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, bio={"content":"

born in 1960. PhD, professor. His research interests include multi-computer systems, embedded systems and applications, and computer control and intelligent systems

"}, bioImg=null, bioContent=

born in 1960. PhD, professor. His research interests include multi-computer systems, embedded systems and applications, and computer control and intelligent systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129447493929, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129317470502, language=CN, stringName=马维华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=5, address=5南京航空航天大学计算机科学与技术学院 南京 211106, bio={"img":"EkOHay4fmPdNtqNTYD6Iaw==","content":"

1960年生. 博士,教授. 主要研究方向为多机系统、嵌入式系统及应用、 计算机控制与智能系统

"}, bioImg=EkOHay4fmPdNtqNTYD6Iaw==, bioContent=

1960年生. 博士,教授. 主要研究方向为多机系统、嵌入式系统及应用、 计算机控制与智能系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019126100439294, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=5, ext=[AuthorCompanyExt(id=1327019126108827903, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126100439294, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016), AuthorCompanyExt(id=1327019126117216512, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126100439294, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5南京航空航天大学计算机科学与技术学院 南京 211106)])]), Author(id=1327019129539768619, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=8, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=bhli@nuaa.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019129611071791, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129539768619, language=EN, stringName=Bohan Li, firstName=Bohan, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106
2Key Laboratory of Brain-Machine Intelligence Technology (Nanjing University of Aeronautics and Astronautics), Ministry of Education, Nanjing 211106
3Key Laboratory of Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 211106, bio={"content":"

born in 1979. PhD, associate professor. Distinguished member of CCF, member of ACM. His main research interests include spatio-temporal database, knowledge graph, and recommendation systems

"}, bioImg=null, bioContent=

born in 1979. PhD, associate professor. Distinguished member of CCF, member of ACM. His main research interests include spatio-temporal database, knowledge graph, and recommendation systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129661403440, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129539768619, language=CN, stringName=李博涵, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1南京航空航天大学人工智能学院 南京 211106
2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106
3软件新技术与产业化协同创新中心重点实验室 南京 211106, bio={"img":"1oO38dZadNp4ocIUvgFnoA==","content":"

1979年生. 博士,副教授. CCF杰出会员,ACM会员. 主要研究方向为时空数据库、知识图谱、推荐系统

"}, bioImg=1oO38dZadNp4ocIUvgFnoA==, bioContent=

1979年生. 博士,副教授. CCF杰出会员,ACM会员. 主要研究方向为时空数据库、知识图谱、推荐系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)]), AuthorCompany(id=1327019125819420917, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=2, ext=[AuthorCompanyExt(id=1327019125827809526, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125819420917, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Brain-Machine Intelligence Technology (Nanjing University of Aeronautics and Astronautics), Ministry of Education, Nanjing 211106), AuthorCompanyExt(id=1327019125836198135, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125819420917, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106)]), AuthorCompany(id=1327019125907501304, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=3, ext=[AuthorCompanyExt(id=1327019125915889913, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125907501304, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Key Laboratory of Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 211106), AuthorCompanyExt(id=1327019125924278522, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125907501304, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3软件新技术与产业化协同创新中心重点实验室 南京 211106)])])], keywords=[Keyword(id=1327019129778843953, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, orderNo=1, keyword=sequential recommendation (SR)), Keyword(id=1327019129850147122, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, orderNo=2, keyword=deformable convolution), Keyword(id=1327019129917255987, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, orderNo=3, keyword=graph neural network), Keyword(id=1327019129967587636, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, orderNo=4, keyword=cold-start), Keyword(id=1327019130026307893, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, orderNo=5, keyword=dynamic convolution), Keyword(id=1327019130093416758, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, orderNo=1, keyword=序列推荐), Keyword(id=1327019130156331319, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, orderNo=2, keyword=可变形卷积), Keyword(id=1327019130223440184, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, orderNo=3, keyword=图神经网络), Keyword(id=1327019130298937657, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, orderNo=4, keyword=冷启动), Keyword(id=1327019130366046522, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, orderNo=5, keyword=动态卷积)], refs=[Reference(id=1327019133796987217, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=0, authorNames=null, journalName=null, refType=null, unstructuredReference=Rendle S, Freudenthaler C, Schmidt-Thieme L. Factorizing personalized Markov chains for next-basket recommendation[C]//Proc of the 19th Int Conf on World Wide Web. New York: ACM, 2010: 811−820, articleTitle=null, refAbstract=null), Reference(id=1327019133864096082, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Beutel A, Faloutsos C. User behavior modeling and fraud detection[J]. IEEE Intelligent Systems, 2016, 31(2): 84−86, articleTitle=null, refAbstract=null), Reference(id=1327019133931204947, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=Donkers T, Loepp B, Ziegler J. Sequential user-based recurrent neural network recommendations[C]//Proc of the 11th ACM Conf on Recommender Systems. New York: ACM, 2017: 152−160, articleTitle=null, refAbstract=null), Reference(id=1327019134015091028, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Hidasi B, Karatzoglou A. Recurrent neural networks with top-k gains for session-based recommendations[C]//Proc of the 27th ACM Int Conf on Information and Knowledge Management. New York: ACM, 2018: 843−852, articleTitle=null, refAbstract=null), Reference(id=1327019134094782805, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Hidasi B, Karatzoglou A, Baltrunas L, et al. Session-based recommendations with recurrent neural networks[J]. arXiv preprint, arXiv: 1511.06939, 2015, articleTitle=null, refAbstract=null), Reference(id=1327019134174474582, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=卢晓凯, 封军, 韩永强, 等. GraphMLP-Mixer: 基于图-多层感知机架构的高效多行为序列推荐方法[J]. 计算机研究与发展, 2024, 61(8): 1917−1929, articleTitle=null, refAbstract=null), Reference(id=1327019134245777751, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Lu Xiaokai, Feng Jun, Han Yongqiang, et al. GraphMLP-Mixer: A graph-MLP architecture for efficient multi-behavior sequential recommendation method[J]. Journal of Computer Research and Development, 2024, 61(8): 1917−1929 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327019134312886616, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=Tang Jiaxi, Wang Ke. Personalized top-n sequential recommendation via convolutional sequence embedding[C]//Proc of the 11th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2018: 565−573, articleTitle=null, refAbstract=null), Reference(id=1327019134371606873, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Yuan Fajie, Karatzoglou A, Arapakis I, et al. A simple convolutional generative network for next item recommendation[C]//Proc of the 12th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2019: 582−590, articleTitle=null, refAbstract=null), Reference(id=1327019134442910042, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=Tanjim M M. DynamicRec: A dynamic convolutional network for next item recommendation[C]//Proc of the 29th ACM Int Conf on Information and Knowledge Management (CIKM−2020). New York: ACM, 2020: 2237−2240, articleTitle=null, refAbstract=null), Reference(id=1327019134501630299, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Chengkang, Julian J. Self-attentive sequential recommendation[C]//Proc of the 18th Int Conf on Data Mining. Piscataway, NJ: IEEE, 2018: 197–206, articleTitle=null, refAbstract=null), Reference(id=1327019134564544860, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jiacheng, Wang Yujie, McAuley J. Time interval aware self-attention for sequential recommendation[C]//Proc of the 13th Int Conf on Web Search and Data Mining. New York: ACM, 2020: 322−330, articleTitle=null, refAbstract=null), Reference(id=1327019134627459421, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Wu Shu, Tang Yuyuan, Zhu Yanqiao, et al. Session-based recommendation with graph neural networks[C]//Proc of the 33rd AAAI Conf on Artificial Intelligence. Piscataway, NJ: IEEE, 2019: 346−353, articleTitle=null, refAbstract=null), Reference(id=1327019134694568286, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=Chang Jianxin, Gao Chen, Zheng Yu, et al. Sequential recommendation with graph neural networks[C]//Proc of the 44th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2021: 378−387, articleTitle=null, refAbstract=null), Reference(id=1327019134778454367, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Zheng Yujia, Liu Siyi, Li Zekun, et al. DGTN: Dual-channel graph transition network for session-based recommendation[C]//Proc of the 20th Int Conf on Data Mining Workshops. Piscataway, NJ: IEEE, 2020: 236−242, articleTitle=null, refAbstract=null), Reference(id=1327019134853951840, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=He Hulingxiao, He Xiangteng, Peng Yuxin, et al. Firzen: Firing strict cold-start items with frozen heterogeneous and homogeneous graphs for recommendation[C]//Proc of IEEE 40th Int Conf on Data Engineering. Piscataway, NJ: IEEE, 2024: 4657−4670, articleTitle=null, refAbstract=null), Reference(id=1327019134929449313, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=Lin Xinyu, Wang Wenjie, Zhao jujia, et al. Temporally and distributionally robust optimization for cold-start recommendation[C]//Proc of the AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024, 38(8), 8750−8758, articleTitle=null, refAbstract=null), Reference(id=1327019135000752482, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Yaqing, Piao Hongming, Dong Daxiang, et al. Warming up cold-start CTR prediction by learning item-specific feature interactions[C]//Proc of the 30th ACM SIGKDD Conf on Knowledge Discovery and Data Mining. New York: ACM, 2024: 3233−3244, articleTitle=null, refAbstract=null), Reference(id=1327019135067861347, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Jianling, Ding Kaize, Caverlee J. Sequential recommendation for cold-start users with meta transitional learning[C]//Proc of the 44th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2021: 1783−1787, articleTitle=null, refAbstract=null), Reference(id=1327019135139164516, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu X, Dong H, Qi L, et al. CMCLR: Cross-modal contrastive learning for user cold-start sequential recommendation[C]//Proc of the 47th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2024: 1589−1598, articleTitle=null, refAbstract=null), Reference(id=1327019135214661989, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Hu Guangneng, Zhang Yu, Yang Qiang. Collaborative cross networks for cross-domain recommendation[C]//Proc of the 27th ACM Int Conf on Information and Knowledge Management. New York: ACM, 2018: 667−676, articleTitle=null, refAbstract=null), Reference(id=1327019135281770854, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Fan Wenqi, Ma Yao, Qing Li, et al. Graph neural networks for social recommendation[C]//Proc of the 26th World Wide Web Conf. New York: ACM, 2019: 417–426, articleTitle=null, refAbstract=null), Reference(id=1327019135340491111, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Chengfeng, Zhao Pengpeng, Liu Yanchi, et al. Graph contextualized self-attention network for session-based recommendation [C]//Proc of the 28th Int Joint Conf on Artificial Intelligence. New York: ACM, 2019: 3940–3946, articleTitle=null, refAbstract=null), Reference(id=1327019135407599976, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Yang Y, Huang C, Xia L, et al. Knowledge graph self-supervised rationalization for recommendation[C]//Proc of the 29th ACM SIGKDD Conf on Knowledge Discovery and Data Mining. New York: ACM, 2023: 3046−3056, articleTitle=null, refAbstract=null), Reference(id=1327019135478903145, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Cui Zeyu, Li Zekun, Wu Shu, et al. Dressing as a whole: Outfit compatibility learning based on node-wise graph neural networks[C]//Proc of the 26th World Wide Web Conf. New York: ACM, 2019: 307–317, articleTitle=null, refAbstract=null), Reference(id=1327019135558594922, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Y, Zhang Y, Zhao Y, et al. Dual Variational graph reconstruction learning for social recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2024, articleTitle=null, refAbstract=null), Reference(id=1327019135625703787, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Khan R H, Miah J, Arafat S M Y, et al. Improving traffic density forecasting in intelligent transportation systems using gated graph neural networks[C]//Proc of the 15th Int Conf on Innovations in Information Technology (IIT). Piscataway, NJ: IEEE, 2023: 104−109, articleTitle=null, refAbstract=null), Reference(id=1327019135697006956, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=Gao Hang, Zhu Xizhou, Lin S, et al. Deformable kernels: Adapting effective receptive fields for object deformation[J]. arXiv preprint, arXiv: 1910.02940, 2019, articleTitle=null, refAbstract=null), Reference(id=1327019135768310125, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang G, Hu Z, Wen G, et al. Dynamic graph convolutional networks by semi-supervised contrastive learning[J]. Pattern Recognition, 2023, 139: 109486, articleTitle=null, refAbstract=null), Reference(id=1327019135843807598, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Liu Yi, Li Bohan, Zang Yalei, et al. A knowledge-aware recommender with attention-enhanced dynamic convolutiona network[C]//Proc of the 30th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2021: 1079−1088, articleTitle=null, refAbstract=null), Reference(id=1327019137529917807, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhao Wayne, Mu Shanlei, Hou Yupeng, et al. RecBole: Towards a unified, comprehensive and efficient framework for recommendation algo-rithms[C]//Proc of the 30th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2021: 4653−4664, articleTitle=null, refAbstract=null), Reference(id=1327019137617998192, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=曾义夫, 牟其林, 周乐, 等. 基于图表示学习的会话感知推荐模型[J]. 计算机研究与发展, 2020, 57(3): 590−603, articleTitle=null, refAbstract=null), Reference(id=1327019137685107057, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Zeng Yifu, Mu Qilin, Zhou Le, et al. Graph embedding based session perception model for next click recommendation[J]. Journal of Computer Research and Development, 2020, 57(3): 590−603 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327019137752215922, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Steffen R, Freudenthaler C, Gantner Z, et al. BPR: Bayesian personalized ranking from implicit feedback[J]. arXiv preprint, arXiv: 1205.2618, 2012, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)]), AuthorCompany(id=1327019125819420917, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=2, ext=[AuthorCompanyExt(id=1327019125827809526, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125819420917, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Brain-Machine Intelligence Technology (Nanjing University of Aeronautics and Astronautics), Ministry of Education, Nanjing 211106), AuthorCompanyExt(id=1327019125836198135, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125819420917, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106)]), AuthorCompany(id=1327019125907501304, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=3, ext=[AuthorCompanyExt(id=1327019125915889913, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125907501304, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Key Laboratory of Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 211106), AuthorCompanyExt(id=1327019125924278522, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125907501304, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3软件新技术与产业化协同创新中心重点实验室 南京 211106)]), AuthorCompany(id=1327019126020747515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=4, ext=[AuthorCompanyExt(id=1327019126029136124, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126020747515, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4Key Laboratory of Advanced Technology for Small and Medium-sized UAVs (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology, Nanjing 210006), AuthorCompanyExt(id=1327019126037524733, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126020747515, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006)]), AuthorCompany(id=1327019126100439294, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=5, ext=[AuthorCompanyExt(id=1327019126108827903, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126100439294, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016), AuthorCompanyExt(id=1327019126117216512, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126100439294, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5南京航空航天大学计算机科学与技术学院 南京 211106)])], figs=[ArticleFig(id=1327019130542207291, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 1, caption=Framework of GISDCN model, figureFileSmall=KBg0gFGq7tAx9y5R79zF8g==, figureFileBig=TbLKWScdUjPQE/mwm/niJw==, tableContent=null), ArticleFig(id=1327019130617704764, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图1, caption=GISDCN模型的框架, figureFileSmall=KBg0gFGq7tAx9y5R79zF8g==, figureFileBig=TbLKWScdUjPQE/mwm/niJw==, tableContent=null), ArticleFig(id=1327019130709979453, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 2, caption=Structure of deformable convolution module, figureFileSmall=gmZbZAGF6nLjSeMJmWYK5g==, figureFileBig=tGklUDpxNjZV/mD/fjqIUQ==, tableContent=null), ArticleFig(id=1327019130768699710, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图2, caption=可变形卷积模块的结构, figureFileSmall=gmZbZAGF6nLjSeMJmWYK5g==, figureFileBig=tGklUDpxNjZV/mD/fjqIUQ==, tableContent=null), ArticleFig(id=1327019130831614271, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 3, caption=Illustration of unfolding sequences, figureFileSmall=0Hawcu4g4eMVMW3unGEVSg==, figureFileBig=rnl9tiWZCvkuiDyjI9KdsA==, tableContent=null), ArticleFig(id=1327019130907111744, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图3, caption=序列的展开示意图, figureFileSmall=0Hawcu4g4eMVMW3unGEVSg==, figureFileBig=rnl9tiWZCvkuiDyjI9KdsA==, tableContent=null), ArticleFig(id=1327019130978414913, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 4, caption=Effect of interpolation module on models, figureFileSmall=0fdUPm4c1pJ6JD78Tqgl/Q==, figureFileBig=vtOl/jNisbEsnEBXgepJaw==, tableContent=null), ArticleFig(id=1327019131058106690, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图4, caption=插值模块对模型的影响, figureFileSmall=0fdUPm4c1pJ6JD78Tqgl/Q==, figureFileBig=vtOl/jNisbEsnEBXgepJaw==, tableContent=null), ArticleFig(id=1327019131116826947, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 5, caption=Influence of different embedding dimensions in GNN and CNN, figureFileSmall=Pgcf9MyK27l8yR5WP5G0aQ==, figureFileBig=owNdjlTMtb772QAoo83srQ==, tableContent=null), ArticleFig(id=1327019131183935812, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图5, caption=GNN和CNN中不同嵌入维度的影响, figureFileSmall=Pgcf9MyK27l8yR5WP5G0aQ==, figureFileBig=owNdjlTMtb772QAoo83srQ==, tableContent=null), ArticleFig(id=1327019131246850373, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 6, caption=Influence of convolution range and proportion of cold start households on the model, figureFileSmall=MPkYMlE7SnkGTPqk7HFFoA==, figureFileBig=YCXB2SOV76UQustGnQQkmQ==, tableContent=null), ArticleFig(id=1327019131305570630, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图6, caption=卷积范围和冷启动用户占比对模型的影响, figureFileSmall=MPkYMlE7SnkGTPqk7HFFoA==, figureFileBig=YCXB2SOV76UQustGnQQkmQ==, tableContent=null), ArticleFig(id=1327019131372679495, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Figure 7, caption=Visualization effect of the model, figureFileSmall=8kbPl9gv8oArAk5P2ivw0A==, figureFileBig=C5toeUTj8YjC61a39hwyNA==, tableContent=null), ArticleFig(id=1327019133042012488, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=图7, caption=模型的可视化效果, figureFileSmall=8kbPl9gv8oArAk5P2ivw0A==, figureFileBig=C5toeUTj8YjC61a39hwyNA==, tableContent=null), ArticleFig(id=1327019133125898569, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Table 1, caption=Statistics of Datasets, figureFileSmall=null, figureFileBig=null, tableContent=
数据集用户数商品数交互数量平均点击数稀疏性
ml-1m603837069595271590.9571
ml-20m23212130701485966640.9893
music10265257356717493650.9890
book364755999442433120.9979
), ArticleFig(id=1327019133188813130, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=表1, caption=

数据集统计信息

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集用户数商品数交互数量平均点击数稀疏性
ml-1m603837069595271590.9571
ml-20m23212130701485966640.9893
music10265257356717493650.9890
book364755999442433120.9979
), ArticleFig(id=1327019133264310603, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Table 2, caption=Comparison of Characteristics Among Different Models, figureFileSmall=null, figureFileBig=null, tableContent=
模型特征交互神经网络CNNGNN序列展开
BPR√××××
FPMC√××××
GRU4REC√√××√
Caser√√√×√
NextItNet√√√×√
DynamicRec√√√×√
SR-GNN√√×√×
GISDCN(本文)√√√√√
), ArticleFig(id=1327019133344002380, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=表2, caption=

不同模型之间的特性比较

, figureFileSmall=null, figureFileBig=null, tableContent=
模型特征交互神经网络CNNGNN序列展开
BPR√××××
FPMC√××××
GRU4REC√√××√
Caser√√√×√
NextItNet√√√×√
DynamicRec√√√×√
SR-GNN√√×√×
GISDCN(本文)√√√√√
), ArticleFig(id=1327019133406916941, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Table 3, caption=Performance of Different Models on Each Dataset, figureFileSmall=null, figureFileBig=null, tableContent=
数据集指标BPRFPMCGRU4RECCaserNextItNetDynamicRecSR-GNNGISDCN (本文)提升/%
注:所有评估指标的最佳结果以黑体数值显示,所有模型的次好评估指标结果以下划线数值突出显示.
bookMRR@100.04560.05010.05220.04540.05730.06630.07400.08008.11
MRR@200.05380.06650.05650.04970.07210.07800.08100.08525.19
NDCG@100.05980.06550.05230.06310.07860.09310.09400.107113.94
NDCG@200.07910.06860.07870.07870.09690.11440.11020.126210.31
HR@100.10990.11630.11910.12190.14900.18110.15440.19568.01
HR@200.17020.16120.18410.18400.22200.26580.21470.27111.99
musicMRR@100.05670.05990.05220.06340.07430.08450.08030.09178.52
MRR@200.06700.06810.05650.07190.08230.09220.08800.09968.03
NDCG@100.07780.08740.05230.09670.10660.11790.11030.12707.72
NDCG@200.08740.11630.06860.12800.13590.14610.13870.15606.78
HR@100.16780.17930.11910.20800.21400.22870.20980.24386.60
HR@200.20930.28340.18410.33210.33040.34070.32250.35895.34
ml-1mMRR@100.05450.05650.05700.06860.07050.06910.08140.08494.30
MRR@200.06190.06190.06410.07550.07740.07740.08650.08993.78
NDCG@100.07890.08040.06600.09580.10110.09750.09260.10584.65
NDCG@200.10000.10010.09320.12090.12440.12790.10890.13092.29
HR@100.15440.15910.16220.18580.19300.19170.15690.19370.36
HR@200.23680.22380.27020.27580.27400.27850.22190.28100.09
ml-20mMRR@100.3890.03700.04570.04570.06170.05110.04320.06698.43
MRR@200.04040.03940.05070.05530.06650.05390.05360.07238.72
NDCG@100.04890.04950.04950.04950.08180.08570.08370.09015.13
NDCG@200.06220.06650.07020.06930.08180.09280.10190.10624.05
HR@100.08760.09130.12330.10070.14860.16650.14840.187412.55
HR@200.14170.12090.19900.16840.22020.22140.22110.248110.76
), ArticleFig(id=1327019133478220110, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=表3, caption=

不同模型在各个数据集上的性能

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集指标BPRFPMCGRU4RECCaserNextItNetDynamicRecSR-GNNGISDCN (本文)提升/%
注:所有评估指标的最佳结果以黑体数值显示,所有模型的次好评估指标结果以下划线数值突出显示.
bookMRR@100.04560.05010.05220.04540.05730.06630.07400.08008.11
MRR@200.05380.06650.05650.04970.07210.07800.08100.08525.19
NDCG@100.05980.06550.05230.06310.07860.09310.09400.107113.94
NDCG@200.07910.06860.07870.07870.09690.11440.11020.126210.31
HR@100.10990.11630.11910.12190.14900.18110.15440.19568.01
HR@200.17020.16120.18410.18400.22200.26580.21470.27111.99
musicMRR@100.05670.05990.05220.06340.07430.08450.08030.09178.52
MRR@200.06700.06810.05650.07190.08230.09220.08800.09968.03
NDCG@100.07780.08740.05230.09670.10660.11790.11030.12707.72
NDCG@200.08740.11630.06860.12800.13590.14610.13870.15606.78
HR@100.16780.17930.11910.20800.21400.22870.20980.24386.60
HR@200.20930.28340.18410.33210.33040.34070.32250.35895.34
ml-1mMRR@100.05450.05650.05700.06860.07050.06910.08140.08494.30
MRR@200.06190.06190.06410.07550.07740.07740.08650.08993.78
NDCG@100.07890.08040.06600.09580.10110.09750.09260.10584.65
NDCG@200.10000.10010.09320.12090.12440.12790.10890.13092.29
HR@100.15440.15910.16220.18580.19300.19170.15690.19370.36
HR@200.23680.22380.27020.27580.27400.27850.22190.28100.09
ml-20mMRR@100.3890.03700.04570.04570.06170.05110.04320.06698.43
MRR@200.04040.03940.05070.05530.06650.05390.05360.07238.72
NDCG@100.04890.04950.04950.04950.08180.08570.08370.09015.13
NDCG@200.06220.06650.07020.06930.08180.09280.10190.10624.05
HR@100.08760.09130.12330.10070.14860.16650.14840.187412.55
HR@200.14170.12090.19900.16840.22020.22140.22110.248110.76
), ArticleFig(id=1327019133549523279, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=EN, label=Table 4, caption=Influence of Deformable Convolution Modules on the Models, figureFileSmall=null, figureFileBig=null, tableContent=
数据集卷积方式MRR@10NDCG@10
注:Baseline2使用通用固定卷积核. 括号内百分数是基于Baseline2的提升值.
bookBaseline20.06530.0872
动态滤波器0.0726(+11.18%)0.0987(+13.19%)
可变形卷积0.0839(+28.48%)0.1088(+24.77%)
ml-1mBaseline20.07130.0963
动态滤波器0.0776(+8.84%)0.1052(+9.24%)
可变形卷积0.0810(+13.6%)0.0810(+13.6%)
), ArticleFig(id=1327019133620826448, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, language=CN, label=表4, caption=

可变形卷积模块对于模型的影响

, figureFileSmall=null, figureFileBig=null, tableContent=
数据集卷积方式MRR@10NDCG@10
注:Baseline2使用通用固定卷积核. 括号内百分数是基于Baseline2的提升值.
bookBaseline20.06530.0872
动态滤波器0.0726(+11.18%)0.0987(+13.19%)
可变形卷积0.0839(+28.48%)0.1088(+24.77%)
ml-1mBaseline20.07130.0963
动态滤波器0.0776(+8.84%)0.1052(+9.24%)
可变形卷积0.0810(+13.6%)0.0810(+13.6%)
)], attaches=null, journal=Journal(id=1324759606947315908, delFlag=0, nameCn=计算机研究与发展, nameEn=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, issn=1000-1239, eissn=null, cn=11-1777/TP, coden=null, periodic=0, language=CN, oaType=null, ccby=null, superviseOffice=null, ownerOffice=null, pubOffice=null, editorOffice=null, officeType=null, aims=null, clcCode=null, officeProv=null, officeCity=null, officeAddr=null, officeZip=null, officeEmail=null, officePhone=null, editDirector=null, officeDirector=null, officeDirectorPhone=null, officeStaffNum=null, officeEmpNum=null, coverPicUrl=k2l8zgqwstuxxevFX2qxVA==, journalPrice=null, startedYear=null, abbrevIsoEn=Journal of Computer Research and Development, journalRemark=null, publicationField=null, createdTime=1793776903141, updatedTime=1793850698612, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Electronic Information, subjectName=null, subjectCodeEn=Electronic Information, subjectNameEn=null, picCn=k2l8zgqwstuxxevFX2qxVA==, picEn=r6k0Ry+RD+c/VXj5QarmDw==, jcr=null, cjcr=null, exts=[JournalExt(id=1325069127710286688, language=CN, name=计算机研究与发展, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698640, updatedTime=1793850698640, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1325069127777395553, language=EN, name=Journal of Computer Research and Development, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1793850698656, updatedTime=1793850698656, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=, submissionAuthorUrl=https://crad.ict.ac.cn/journalx/authorLogOn.action, submissionEditorUrl=https://crad.ict.ac.cn/journalx/editorLogOn.action, submissionReviewUrl=https://crad.ict.ac.cn/journalx/expertLogOn.action, submissionCeEditorUrl=https://crad.ict.ac.cn/journalx/editorInChiefLogOn.action, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1324759914146529478, websiteList=[Website(id=1325070300139892771, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN, language=CN, createTime=1793850978169, createBy=18614031015, updateTime=1793851040259, updateBy=18614031015, name=计算机研究与发展-中文, tplId=1146099689490845704, title=计算机研究与发展, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070893751349360, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=articleTextType, value=kx, createTime=1793851119697, updateTime=1793851119697, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893730377837, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=banner, value=null, createTime=1793851119692, updateTime=1793851119692, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893772320883, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=grayFlag, value=0, createTime=1793851119702, updateTime=1793851119702, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893721989228, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851119690, updateTime=1793851119690, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893784903797, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=minRunFlag, value=0, createTime=1793851119705, updateTime=1793851119705, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893742960751, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/file/pic, createTime=1793851119695, updateTime=1793851119695, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893780709492, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=silenceFlag, value=0, createTime=1793851119704, updateTime=1793851119704, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893738766446, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1793851119694, updateTime=1793851119694, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893755543665, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeColor, value=null, createTime=1793851119698, updateTime=1793851119698, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070893763932274, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300139892771, code=themeStyle, value=null, createTime=1793851119700, updateTime=1793851119700, creator=18614031015, updator=18614031015)]), Website(id=1325070300580294698, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1324759914146529478, journalNameCn=null, journalNameEn=null, grayFlag=null, tenantId=1146029695717560320, platformId=null, journalGroupId=null, journalGroupNameCn=null, journalGroupNameEn=null, type=1, domain=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN, language=EN, createTime=1793850978274, createBy=18614031015, updateTime=1793851036638, updateBy=18614031015, name=计算机研究与发展-英文, tplId=1146101810881728533, title=Journal of Computer Research and Development, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1325070925288321148, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=articleTextType, value=kx, createTime=1793851127216, updateTime=1793851127216, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925263155321, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=banner, value=null, createTime=1793851127210, updateTime=1793851127210, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925309292671, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=grayFlag, value=0, createTime=1793851127221, updateTime=1793851127221, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925254766712, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=logo, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic?fileId=U27tDsy717YF7XSPA+Vwdw==, createTime=1793851127208, updateTime=1793851127208, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925326069889, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=minRunFlag, value=0, createTime=1793851127225, updateTime=1793851127225, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925279932539, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/file/pic, createTime=1793851127214, updateTime=1793851127214, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925317681280, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=silenceFlag, value=0, createTime=1793851127223, updateTime=1793851127223, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925271543930, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1793851127212, updateTime=1793851127212, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925292515453, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeColor, value=null, createTime=1793851127217, updateTime=1793851127217, creator=18614031015, updator=18614031015), WebsiteProps(id=1325070925300904062, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1325070300580294698, code=themeStyle, value=null, createTime=1793851127219, updateTime=1793851127219, creator=18614031015, updator=18614031015)])], journalTitle=计算机研究与发展, weixinUrl=null, journalUrl=https://crad.ict.ac.cn/, iacademicId=null, status=1, seqNo=null, journalTitleEn=Journal of Computer Research and Development, journalPhotoCn=k2l8zgqwstuxxevFX2qxVA==, journalPhotoEn=r6k0Ry+RD+c/VXj5QarmDw==, journalFirstLetter=J, journalRecommend=null, journalNew=null, journalCollection=null, jcrJf=null, cjcrJf=null, jcrJfStr=null, cjcrJfStr=null, submissionFirstDecision=null, sciSubjectClassification=null, casSubjectClassification=null, citeScore=null, totalCitationFrequency=null, icpCode=null, psCode=null, advertisingLicenseCode=null, copyrightInformation=null, country=null, option=, provinceCode=null, provinceName=null, collectFlag=false, interPubPlatform=, interPubPlatformUrl=null), detailUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202220749, detailUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202220749, pdfUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/PDF/10.7544/issn1000-1239.202220749, pdfUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/PDF/10.7544/issn1000-1239.202220749, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1759248000000, fullTextJson=null, articleText=null, reference=null)
收藏切换
基于图插值和可变形卷积网络的序列推荐
刘昕悦 1 , 尹海莲 4 , 臧亚磊 1 , 吴文隆 1 , 卓俊男 1 , 徐凤如 1 , 陈吕莹 1 , 马维华 5 , 李博涵 1, 2, 3
计算机研究与发展 | 人工智能 2025,62(10): 2583-2594
收起
收藏切换
计算机研究与发展 |人工智能 2025 , 62 (10) : 2583 -2594
基于图插值和可变形卷积网络的序列推荐
全屏
[Author(id=1327019126175936770, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126247239940, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126175936770, language=EN, stringName=Xinyue Liu, firstName=Xinyue, middleName=null, lastName=Liu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

"}, bioImg=null, bioContent=

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019126326931717, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126175936770, language=CN, stringName=刘昕悦, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"hv9eUo+PLu0dbaQtB71RRA==","content":"

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

"}, bioImg=hv9eUo+PLu0dbaQtB71RRA==, bioContent=

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019126402429191, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126473732361, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126402429191, language=EN, stringName=Hailian Yin, firstName=Hailian, middleName=null, lastName=Yin, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=4, address=4Key Laboratory of Advanced Technology for Small and Medium-sized UAVs (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology, Nanjing 210006, bio={"content":"

born in 1979. PhD, associate professor. Her main research interests include multidisciplinary design optimization and machine learning

"}, bioImg=null, bioContent=

born in 1979. PhD, associate professor. Her main research interests include multidisciplinary design optimization and machine learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019126540841226, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126402429191, language=CN, stringName=尹海莲, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=4, address=4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006, bio={"img":"uUl4gZx//Y05SdHXvI4txw==","content":"

1979年生. 博士,副研究员. 主要研究方向为多学科设计优化、机器学习

"}, bioImg=uUl4gZx//Y05SdHXvI4txw==, bioContent=

1979年生. 博士,副研究员. 主要研究方向为多学科设计优化、机器学习

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019126020747515, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=4, ext=[AuthorCompanyExt(id=1327019126029136124, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126020747515, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4Key Laboratory of Advanced Technology for Small and Medium-sized UAVs (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology, Nanjing 210006), AuthorCompanyExt(id=1327019126037524733, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126020747515, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006)])]), Author(id=1327019126612144396, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126687641870, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126612144396, language=EN, stringName=Yalei Zang, firstName=Yalei, middleName=null, lastName=Zang, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 1996. Master candidate. Student member of CCF. His main research interests include sequential recommendation and graph neural networks

"}, bioImg=null, bioContent=

born in 1996. Master candidate. Student member of CCF. His main research interests include sequential recommendation and graph neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019126771527951, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126612144396, language=CN, stringName=臧亚磊, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"4WlZ15tnNpA0m2lAtF6obg==","content":"

1996年生. 硕士研究生. CCF学生会员. 主要研究方向为序列推荐、 图神经网络

"}, bioImg=4WlZ15tnNpA0m2lAtF6obg==, bioContent=

1996年生. 硕士研究生. CCF学生会员. 主要研究方向为序列推荐、 图神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019126834442513, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=3, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019126905745683, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126834442513, language=EN, stringName=Wenlong Wu, firstName=Wenlong, middleName=null, lastName=Wu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2001. Master candidate. Student member of CCF. His main research interests include large language model and recommendation systems

"}, bioImg=null, bioContent=

born in 2001. Master candidate. Student member of CCF. His main research interests include large language model and recommendation systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019128591855893, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019126834442513, language=CN, stringName=吴文隆, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"8WK03dFG1iDz2SBDeToL0Q==","content":"

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为大语言模型、推荐系统

"}, bioImg=8WK03dFG1iDz2SBDeToL0Q==, bioContent=

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为大语言模型、推荐系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019128709296407, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=4, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019128780599577, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128709296407, language=EN, stringName=Junnan Zhuo, firstName=Junnan, middleName=null, lastName=Zhuo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

"}, bioImg=null, bioContent=

born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019128843514138, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128709296407, language=CN, stringName=卓俊男, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"dFoSOgoJN8GJlunF1GVtMw==","content":"

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

"}, bioImg=dFoSOgoJN8GJlunF1GVtMw==, bioContent=

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019128902234396, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=5, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019128981926174, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128902234396, language=EN, stringName=Fengru Xu, firstName=Fengru, middleName=null, lastName=Xu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 2001. Master candidate. Student member of CCF. Her main research interests include recommendation systems and large language models

"}, bioImg=null, bioContent=

born in 2001. Master candidate. Student member of CCF. Her main research interests include recommendation systems and large language models

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129044840735, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019128902234396, language=CN, stringName=徐凤如, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"8v45a+G3QHfVWIWmrLuINg==","content":"

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、大语言模型

"}, bioImg=8v45a+G3QHfVWIWmrLuINg==, bioContent=

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、大语言模型

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019129120338209, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=6, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019129191641379, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129120338209, language=EN, stringName=lüying Chen, firstName=lüying, middleName=null, lastName=Chen, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, bio={"content":"

born in 1998. Master candidate. Student member of CCF. Her main research interests include cross-domain recommendation and sequential recommendation

"}, bioImg=null, bioContent=

born in 1998. Master candidate. Student member of CCF. Her main research interests include cross-domain recommendation and sequential recommendation

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129250361636, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129120338209, language=CN, stringName=陈吕莹, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, address=1南京航空航天大学人工智能学院 南京 211106, bio={"img":"HQogqnbpsi5m6luJlK6dtA==","content":"

1998年生. 硕士研究生. CCF学生会员. 主要研究方向为跨域推荐、序列推荐

"}, bioImg=HQogqnbpsi5m6luJlK6dtA==, bioContent=

1998年生. 硕士研究生. CCF学生会员. 主要研究方向为跨域推荐、序列推荐

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)])]), Author(id=1327019129317470502, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=7, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019129376190760, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129317470502, language=EN, stringName=Weihua Ma, firstName=Weihua, middleName=null, lastName=Ma, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=5, address=5College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, bio={"content":"

born in 1960. PhD, professor. His research interests include multi-computer systems, embedded systems and applications, and computer control and intelligent systems

"}, bioImg=null, bioContent=

born in 1960. PhD, professor. His research interests include multi-computer systems, embedded systems and applications, and computer control and intelligent systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129447493929, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129317470502, language=CN, stringName=马维华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=5, address=5南京航空航天大学计算机科学与技术学院 南京 211106, bio={"img":"EkOHay4fmPdNtqNTYD6Iaw==","content":"

1960年生. 博士,教授. 主要研究方向为多机系统、嵌入式系统及应用、 计算机控制与智能系统

"}, bioImg=EkOHay4fmPdNtqNTYD6Iaw==, bioContent=

1960年生. 博士,教授. 主要研究方向为多机系统、嵌入式系统及应用、 计算机控制与智能系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019126100439294, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=5, ext=[AuthorCompanyExt(id=1327019126108827903, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126100439294, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016), AuthorCompanyExt(id=1327019126117216512, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019126100439294, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=5南京航空航天大学计算机科学与技术学院 南京 211106)])]), Author(id=1327019129539768619, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, orderNo=8, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=bhli@nuaa.edu.cn, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327019129611071791, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129539768619, language=EN, stringName=Bohan Li, firstName=Bohan, middleName=null, lastName=Li, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106
2Key Laboratory of Brain-Machine Intelligence Technology (Nanjing University of Aeronautics and Astronautics), Ministry of Education, Nanjing 211106
3Key Laboratory of Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 211106, bio={"content":"

born in 1979. PhD, associate professor. Distinguished member of CCF, member of ACM. His main research interests include spatio-temporal database, knowledge graph, and recommendation systems

"}, bioImg=null, bioContent=

born in 1979. PhD, associate professor. Distinguished member of CCF, member of ACM. His main research interests include spatio-temporal database, knowledge graph, and recommendation systems

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327019129661403440, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, authorId=1327019129539768619, language=CN, stringName=李博涵, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=1, 2, 3, address=1南京航空航天大学人工智能学院 南京 211106
2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106
3软件新技术与产业化协同创新中心重点实验室 南京 211106, bio={"img":"1oO38dZadNp4ocIUvgFnoA==","content":"

1979年生. 博士,副教授. CCF杰出会员,ACM会员. 主要研究方向为时空数据库、知识图谱、推荐系统

"}, bioImg=1oO38dZadNp4ocIUvgFnoA==, bioContent=

1979年生. 博士,副教授. CCF杰出会员,ACM会员. 主要研究方向为时空数据库、知识图谱、推荐系统

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327019125731340530, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=1, ext=[AuthorCompanyExt(id=1327019125739729139, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106), AuthorCompanyExt(id=1327019125748117748, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125731340530, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=1南京航空航天大学人工智能学院 南京 211106)]), AuthorCompany(id=1327019125819420917, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=2, ext=[AuthorCompanyExt(id=1327019125827809526, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125819420917, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2Key Laboratory of Brain-Machine Intelligence Technology (Nanjing University of Aeronautics and Astronautics), Ministry of Education, Nanjing 211106), AuthorCompanyExt(id=1327019125836198135, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125819420917, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106)]), AuthorCompany(id=1327019125907501304, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, xref=3, ext=[AuthorCompanyExt(id=1327019125915889913, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125907501304, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3Key Laboratory of Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 211106), AuthorCompanyExt(id=1327019125924278522, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327018072738738527, companyId=1327019125907501304, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=3软件新技术与产业化协同创新中心重点实验室 南京 211106)])])]
刘昕悦1, 尹海莲4, 臧亚磊1, 吴文隆1, 卓俊男1, 徐凤如1, 陈吕莹1, 马维华5, 李博涵1, 2, 3
作者信息
  • 1南京航空航天大学人工智能学院 南京 211106
  • 2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106
  • 3软件新技术与产业化协同创新中心重点实验室 南京 211106
  • 4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006
  • 5南京航空航天大学计算机科学与技术学院 南京 211106
作者简介:

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

1979年生. 博士,副研究员. 主要研究方向为多学科设计优化、机器学习

1996年生. 硕士研究生. CCF学生会员. 主要研究方向为序列推荐、 图神经网络

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为大语言模型、推荐系统

2000年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、图神经网络

2001年生. 硕士研究生. CCF学生会员. 主要研究方向为推荐系统、大语言模型

1998年生. 硕士研究生. CCF学生会员. 主要研究方向为跨域推荐、序列推荐

1960年生. 博士,教授. 主要研究方向为多机系统、嵌入式系统及应用、 计算机控制与智能系统

1979年生. 博士,副教授. CCF杰出会员,ACM会员. 主要研究方向为时空数据库、知识图谱、推荐系统

A Graph-Based Interpolation Sequential Recommender with Deformable Convolutional Network
Xinyue Liu1, Hailian Yin4, Yalei Zang1, Wenlong Wu1, Junnan Zhuo1, Fengru Xu1, lüying Chen1, Weihua Ma5, Bohan Li1, 2, 3
Affiliations
  • 1College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics, Nanjing 211106
  • 2Key Laboratory of Brain-Machine Intelligence Technology (Nanjing University of Aeronautics and Astronautics), Ministry of Education, Nanjing 211106
  • 3Key Laboratory of Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing 211106
  • 4Key Laboratory of Advanced Technology for Small and Medium-sized UAVs (Nanjing University of Aeronautics and Astronautics), Ministry of Industry and Information Technology, Nanjing 210006
  • 5College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 210016
  • born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

    born in 1979. PhD, associate professor. Her main research interests include multidisciplinary design optimization and machine learning

    born in 1996. Master candidate. Student member of CCF. His main research interests include sequential recommendation and graph neural networks

    born in 2001. Master candidate. Student member of CCF. His main research interests include large language model and recommendation systems

    born in 2000. Master candidate. Student member of CCF. Her main research interests include recommendation systems and graph neural networks

    born in 2001. Master candidate. Student member of CCF. Her main research interests include recommendation systems and large language models

    born in 1998. Master candidate. Student member of CCF. Her main research interests include cross-domain recommendation and sequential recommendation

    born in 1960. PhD, professor. His research interests include multi-computer systems, embedded systems and applications, and computer control and intelligent systems

    born in 1979. PhD, associate professor. Distinguished member of CCF, member of ACM. His main research interests include spatio-temporal database, knowledge graph, and recommendation systems

出版时间: 2025-10-01 doi: 10.7544/issn1000-1239.202220749
文章导航
收藏切换

序列推荐系统(sequential recommendation system,SRS)旨在基于用户的历史行为偏好预测下一步行为. 尽管针对序列推荐提出了许多有效的方法,但仍然存在根本性的挑战. 首先,随着在线服务的普及,推荐系统需要同时服务于热启动用户和冷启动用户. 然而,由于难以从交互数据有限的序列中学习到有效的序列依赖关系,大多数依赖于用户-项目交互的现有模型失去了优势. 其次,由于现实中用户意图的可变性和主观随机性,用户在其历史序列中的行为往往是隐含和复杂的,很难从这些用户-项目交互数据中捕获这种动态转变模式. 提出了一种基于图插值和可变形卷积网络的序列推荐(graph-based interpolation sequential recommender with deformable convolutional network,GISDCN)模型. 对于冷启动用户,将序列对象重新构建成图,并提取全局序列中的知识来推断用户可能的偏好. 为了捕捉复杂的顺序依赖关系,使用可变形卷积网络来生成更健壮和灵活的卷积核. 最后,在4个数据集上进行了综合实验,验证了模型的有效性. 实验结果表明,GISDCN优于大多数主流的模型.

序列推荐  /  可变形卷积  /  图神经网络  /  冷启动  /  动态卷积

Sequential recommendation systems aim to predict users’ next actions based on the preferences learned from their historical behaviors. There are still fundamental challenges for sequential recommender. First, with the popularization of online services, recommender need serve both the warm-start and cold-start users. However, the most existing models depending on user-item interactions lose merits due to the difficulty of learning sequential dependencies with limited interactions. Second, users’ behaviors in their historical sequences are often implicit and complex due to the objective variability of reality and the subjective randomness of users’ intentions. It is difficult to capture the dynamic transition patterns from these user-item interactions. In this work, we propose a graph-based interpolation sequential recommender with deformable convolutional network (GISDCN). For cold-start users, we reconstruct item sequences into a graph to infer users’ possible preferences. To capture the complex sequential dependencies, we employ the deformable convolutional network to generate more robust and flexible filters. Finally, we conduct comprehensive experiments and verify the effectiveness of our model. The experimental results demonstrate that GISDCN outperforms most of the state-of-the-art models at cold-start conditions.

sequential recommendation (SR)  /  deformable convolution  /  graph neural network  /  cold-start  /  dynamic convolution
刘昕悦, 尹海莲, 臧亚磊, 吴文隆, 卓俊男, 徐凤如, 陈吕莹, 马维华, 李博涵. 基于图插值和可变形卷积网络的序列推荐. 计算机研究与发展, 2025 , 62 (10) : 2583 -2594 . DOI: 10.7544/issn1000-1239.202220749
Xinyue Liu, Hailian Yin, Yalei Zang, Wenlong Wu, Junnan Zhuo, Fengru Xu, lüying Chen, Weihua Ma, Bohan Li. A Graph-Based Interpolation Sequential Recommender with Deformable Convolutional Network[J]. Journal of Computer Research and Development, 2025 , 62 (10) : 2583 -2594 . DOI: 10.7544/issn1000-1239.202220749
推荐系统(recommendation system,RS)旨在学习用户的偏好并为用户选择有用的信息. 随着数据的快速爆炸式增长,推荐系统对于缓解信息过载问题变得越来越重要. 与将用户的一般偏好表示为静态模式的大多数现有推荐系统不同,序列推荐系统(sequential recommendation system,SRS)从用户的交互序列捕获动态依赖关系. 为了更好地满足用户的需求,SRS近年来越来越受到关注.
为了捕捉顺序依赖,早期方法如FPMC[1]引入马尔可夫链来捕获顺序模式. 然后Fossil[2]通过组合Matrix因式分解和马尔可夫链同时适应长期和短期依赖. 然而,基于马尔可夫链的模型不能从序列中提取高阶依赖项. 最近,基于神经网络的方法因其强大的表示能力,已被广泛用于序列依赖项的建模,这些方法采用了不同的网络结构,包括基于循环神经网络(recurrent neural network,RNN)[3-6]、卷积神经网络(convolutional neural network,CNN)[7-9]、注意力机制[10-11]和图神经网络(graph neural network,GNN)[12-14]的模型. 尽管现有的研究工作已经取得了优良的表现,但序列推荐仍然存在根本性的挑战. 首先,由于大多数序列推荐模型主要使用用户-物品来学习用户的偏好,这些模型的性能依赖于充足的历史交互数据. 但是,随着在线服务的进一步普及,越来越多的平台在面对没有足够历史数据的新用户的时候很难从只有几个历史数据的短序列中提取序列依赖关系并推断它们的偏好. 因此,冷启动用户问题仍然是序列推荐所面临的根本挑战. 此外,从序列中提取的偏好总是满足一些实际的用户特殊时期的需求. 由于现实的多变性和用户意图的随机性,包含在序列中的这些偏好往往难以提取. 具体来说,复杂的序列依赖包括许多不同的模式,例如点级、联合级和跳级模式[7]. 同时,不同类型的物品对于预测有不同的贡献,有些物品甚至是干扰下一点击行为预测的噪声.
为了克服这些挑战,不同算法被提出. 对于冷启动用户,现有模型大多利用用户和物品属性等辅助信息来弥补信息的不足[15]. 为了有效应对有限的数据并进行用户跨平台迁移学习,一些方法[16-17]进一步将辅助信息与元路径结合起来以缓解冷启动问题. 为摆脱额外的辅助知识,MetaTL[18]通过元学习建模用户的转换模式,并为只有有限交互数据的冷启动用户提供快速学习. 然而,文献[18]方法几乎没有考虑热启动场景中的长期序列依赖和全局序列中用户的核心偏好. 对于复杂的序列依赖关系,也有许多不同的方法对其进行捕捉. 基于CNN的方法采用卷积运算,但受限于固定卷积核[8]. 基于RNN的模型受到强顺序假设的限制,不能灵活适应复杂多变的数据. 基于注意力的模型[17-18]采用自注意力机制以区分物品在序列中不同的重要性. 然而,这些模型中的物品之间的点对点交互很难表示复杂的联级模式并编码位置信息. 因此,有必要引入一种新的顺序推荐器,用于提取项目之间的复杂依赖关系,并有能力满足冷启动[19]用户的需求. 本文提出了一种新颖的序列推荐模型GISDCN(graph-based interpolation sequential recommender with deformable convolutional network),结构如图1所示,该模型改进了复杂序列依赖关系的提取方式并降低了针对冷启动用户偏好建模的难度. 以缓解冷启动用户的问题,本文利用来自全局序列的热启动用户的信息来推断冷启动用户的可能偏好. 本文将序列重新构建为全局图,并提取核心兴趣节点的相邻节点,以构成这些短序列中的交互. 为捕捉复杂的序列依赖关系,模型采用了可变形卷积网络. 使用偏移量来根据当前输入生成鲁棒内核,以提高卷积的灵活性. 因此,本文的模型可以有效区分不同物品在预测中的影响. 最后,使用4个数据集来评估本文提出的模型. 结果表明,GISDCN比主流的方法表现更好,尤其是在冷启动的情况下. 本文的贡献有3点:
1)提出了一种基于图神经网络的插值方法,旨在从全局序列中推断用户的潜在偏好,从而有效地缓解冷启动用户问题.
2)对用于序列推荐的动态卷积网络进行了改进,采用可变形卷积网络以增强模型从序列中提取复杂依赖关系的能力.
3)在4个公共数据集上进行的实验验证了所提模型在热启动和冷启动条件下的有效性. 同时,设计了详尽的消融实验,以分析各个模块的贡献及其相互作用机制.
序列推荐系统旨在根据用户的历史交互来预测用户的下一步行为. 一些早期的方法,如FPMC[1]和Fossil,使用马尔可夫链来建模序列中对象之间的关系. 然而,基于马尔可夫链的方法受限于模型的复杂性,难以提取高阶依赖. 然后神经网络被引入作为序列数据建模的典型方法来提取用户的偏好,基于 RNN的模型[4-5]用于从用户交互数据中提取序列模式. 这些模型都是基于相邻对象之间的强顺序假设,之后基于CNN的模型[7-9]被提出并在复杂的序列模式建模方面取得了良好的性能[7]. 由于在推荐场景中可以将序列数据重构为图构,因此基于GNN的模型成为近年来流行的方法. 例如,SR-GNN[10]将序列建模为图形结构数据并捕获物品的复杂转换. SURGE[11]将松散的项目序列重构为紧密的项目-项目兴趣图,并从嘈杂的行为序列中提取用户的兴趣.
本文主要采用GNN来捕捉用户的核心兴趣,并结合CNN的优点提取序列中的动态转换偏好信息.
鲁棒能力强的推荐系统需要具备在用户没有足够历史交互数据下对其进行建模的能力. 常见的解决方案是利用辅助信息,Firzen[15]使用新项目的描述,基于异构图与同构图学习的推荐方法,以缓解冷启动项目的问题. CoNet[20]使用神经网络作为基础模型,并从相关领域迁移知识以缓解冷启动问题. 此外将元学习用于用户迁移学习也可以缓解冷启动问题. DRO[16]提出一种基于分布鲁棒优化的新型方法. EmerG [17]通过学习项目特定的特征交互模式来预测冷启动点击率的新方法. MetaTL[18]摆脱了额外的辅助信息,并通过元学习从很少的用户-项目交互中有效地捕获数据中的短程过渡状态. 本文采用GNN来聚合用户的核心偏好和来自全局序列的长期兴趣. 根据用户可能的兴趣点,在这些短序列中组成交互,以缓解冷启动问题.
由于CNN从复杂数据中提取高阶特征的能力很强,因此最初用于计算机视觉(computer vision,CV)任务. 最近,用于序列推荐的CNN引起了人们的关注. Caser[7]使用水平和垂直卷积过滤器来学习用户的复杂行为,例如点级、联合级和跳过行为. 为学习来自短期和长期序列依赖的更加丰富的表示,Yuan等人[8]采用带有空洞的卷积层,并引入了一个简单但有效的生成模型. 由于用户的偏好和需求本质上是动态的,Tanjim等人[9]提出了DynamicRec模型,采用当前动态输入方式计算卷积核. 在本文工作中,设计了更强大的过滤器,进一步提高了动态卷积网络中卷积核的灵活性,以提取复杂的序列依赖关系.
GNN将神经网络技术应用于图数据. 利用GNN在学习图信息表示方面的优势,一些推荐系统已经使用GNN来解决传统序列推荐面临的挑战. 从模型的角度来看,基于GNN的推荐系统主要分为3类:基于图注意力网络(graph attention network,GAT)的推荐系统、基于门控图神经网络(gated graph neural network,GGNN)的推荐系统,以及基于图卷积网络(graph convolutional network,GCN)的推荐系统. GAT是将注意力机制引入GNN,以侧重学习图中用节点代表的用户或者商品之间的联系,从而更加精确地获得它们之间的关系. 在这种情况下,更重要的用户或物品,或者特定用户或物品的影响被加重,这更符合现实世界的情况,而且已被证明有利于推荐系统. 此类方法被广泛用于基于社交图的推荐[21],如会话的推荐[22]和基于知识图谱的推荐[23]. GGNN是将门控循环单元(gated recurrent unit,GRU)引入GNN,通过迭代吸收图中其他节点的影响来学习优化的节点表示,从而全面捕捉节点间的关系. GGNN广泛用于对序列图中项目之间的复杂转换进行建模,以进行序列推荐,如Wu等人[12]将用户的会话构建为连接的图结构,再使用门控机制进行节点信息的传播,但是该方法只考虑局部的图数据,而没有利用用户和商品的全局信息. Cui等人[24]对不同类别的物品之间的复杂交互进行建模以进行时尚推荐,并且取得了不错的推荐性能. GCN通常利用图结构和节点特征信息来学习如何从本地图邻居节点迭代地聚合特征信息. 具体来讲,它是通过利用卷积和池化操作有效地聚合来自图中邻域的信息以学习用户和项目的信息嵌入. 由于其强大的特征提取和学习能力,特别是结合图结构和节点内容信息的能力,GCN被广泛应用于各种基于图的推荐系统,如使用GCN 的社交推荐[25]. 卢晓凯等人[6]通过构造全局物品图来增强模型对序列间协同信号的建模,将感知机-混合器架构与GCN结合对用户兴趣进行挖掘,但是它们多是利用外部的信息去构建图,而对于没有外部数据的商品是不友好的.
本文中使用的符号和给出序列推荐任务的正式定义如下:
$ V=\left\{{v}_{1},{v}_{2},… ,{v}_{\left|V\right|}\right\} $和$ U=\left\{{u}_{1},{u}_{2},… ,{u}_{\left|U\right|}\right\} $分别定义为交互序列集合和用户集合,${S} $是序列集合. 对于用户的点击序列$ u=\left\{{v}_{1}^{u},{v}_{2}^{u},… ,{v}_{n-1}^{u}\right\} $,n是序列的长度,序列推荐的主要任务是基于序列偏好来预测最后的点击对象$ {v}_{n} $.
输入:每一个用户的历史点击序列;
输出:通过对用户过往的交互行为进行建模,预测用户在给定环境下最可能选择的下一个交互目标.
本文所提出的模型GISDCN利用GNN为冷启动用户进行序列补全,并采用可变形卷积网络来提取序列中的依赖关系.
为了缓解序列中点击记录的不足,本文尝试利用所有用户的核心兴趣,对这些短序列进行补全. 使用 GNN来找到核心兴趣节点,并将合适的物品与候选节点区分开来,以适应用户在短序列中可能的偏好.
首先,将每个序列根据其长度l进行划分. 长序列用于构建全局图,短序列需要进行插值. 然后构建有向全局图:$ {G}_{\mathrm{g}}=\left({V}_{\mathrm{g}},{\varepsilon }_{\mathrm{g}}\right) $. 每个节点$ {v}_{i}\in {V}_{\mathrm{g}} $对应一个点击的对象,每条边$ \left({v}_{i-1},{v}_{i}\right)\in {\varepsilon }_{\mathrm{g}} $表示用户在$ {v}_{i-1} $之后的点击对象$ {v}_{i} $. 本文用边的出现频次除以边起始节点的出度来区分边的重要性.
在构建全局图之后,本文展示了如何对项目转换进行编码并通过图神经网络学习节点表示. 利用GGNNs[26]来学习节点的表示,具体来说,$({\boldsymbol{h}}_{1},{\boldsymbol{h}}_{2},… , {\boldsymbol{h}}_{n}) $中$\boldsymbol h_i $是维度为d的嵌入向量,n是序列的长度. 那么对于每个长序列,所有节点的学习过程可以表示为[5,27]
$ \boldsymbol{h}'_{i}=GGNNs\left(({\boldsymbol{h}}_{1},{\boldsymbol{h}}_{2},… ,{\boldsymbol{h}}_{n}),{\boldsymbol{A}}_{{s}}\right), $
其中,$ {\boldsymbol{A}}_{{S}} $是描述节点之间连接的矩阵,计算函数如下:
$\begin{split} {\boldsymbol{A}}_{{S}}=\;&\left(\left(\begin{array}{ccc}{q}_{11}& … & {q}_{1n}\\\vdots & & \vdots \\{q}_{1n}& … & {q}_{nn}\end{array}\right)\left(\begin{array}{c}\dfrac{1}{{q}_{1}^{\text{out}}}\\\vdots \\\dfrac{1}{{q}_{n}^{\text{out}}}\end{array}\right)\right)\Bigg\|\\&\left({\left(\begin{array}{ccc}{q}_{11}& … & {q}_{1n}\\\vdots & & \vdots \\{q}_{1n}& … & {q}_{nn}\end{array}\right)}^{\mathrm{T}}\left(\begin{array}{c}\dfrac{1}{{q}_{1}^{\text{in}}}\\\vdots \\\dfrac{1}{{q}_{n}^{\text{in}}}\end{array}\right)\right),\end{split} $
其中,$ \left|\right| $表示连接操作,$ {q}_{ij} $代表节点$ {v}_{i} $和$ {v}_{j} $之间的边的数量,$ {q}_{i}^{\mathrm{o}\mathrm{u}\mathrm{t}} $和$ {q}_{i}^{\mathrm{i}\mathrm{n}} $分别表示节点的出度和入度. 在对图上所有的节点进行表示学习后,$ \boldsymbol{h}'_i $代表聚合图上邻居节点信息后新的表示.
在给定的全局图基础上,本文提出了一种方法,该方法通过核心兴趣节点对与冷启动用户相对应的短序列进行补全. 每个短序列中拥有最大的度数值的节点作为用户的核心兴趣节点. 在由所有长序列组成的全局图中,将不同序列的信息与核心兴趣节点进行聚合,以表示核心节点. 为了减少核心节点表示过程中的噪声,将只考虑核心兴趣点邻近的ε项:
$ {s}_{j}=\left\{{v}_{i}\mid {v}_{{i}'}={v}_{c}\in {s}_{\mathrm{o}\mathrm{r}\mathrm{i}},i\in \left[{i}'-\varepsilon ,{i}'+\varepsilon \right]\right\}, $
其中,$ {s}_{\mathrm{o}\mathrm{r}\mathrm{i}} $表示序列,而每一个序列的表示可以通过注意力机制得到:
$ {\boldsymbol{H}}_{{s}_{j}}=\sum_{{v}_{i}\in {s}_{j}}\left({\boldsymbol{W}}_{3}^{\mathrm{T}}\sigma \left({\boldsymbol{W}}_{1}{\boldsymbol{h}}_{{v}_{\text{last}}}'+{\boldsymbol{W}}_{2}{\boldsymbol{h}}_{{v}_{i}}'+\boldsymbol{b}\right)\right){\boldsymbol{h}}_{{v}_{i}}'\text{,} $
其中,$ {\boldsymbol{W}} $是权重,$ \boldsymbol{b} $是偏置,$ {v}_{\text{last}} $是序列$ {s}_{j} $中的最后一个被点击对象.
为了区分核心兴趣节点在不同序列中的重要性并获得更加鲁棒的表示,为序列$ {s}_{j} $的每个嵌入表示分配一个权重. $ {\boldsymbol{H}}_{{s}_{j}} $是没有核心节点的序列表示. 然后,通过添加不同的序列表示来计算核心兴趣节点的表示. 最后,选择全局图中与核心兴趣节点相邻的节点作为候选节点,并通过将其表示乘以代表核心兴趣节点的表示来获得最终分数:
$ {score}_{i}=\boldsymbol{h}'^{\rm T}_{i}\left(\sum _{{s}_{j}\in {S}_{\mathrm{a}\mathrm{g}\mathrm{g}}}\left(1-\mathrm{cos}\left(\boldsymbol{W}\odot \boldsymbol{H}_{{s}_{j}},\boldsymbol{W}\odot \boldsymbol{H}'_{{s}_{j}}\right)\right){\boldsymbol{H}}_{{s}_{j}}\right), $
其中,得分最高的N个对象将被选择作为短序列插值的节点.
在对短序列进行插值之后,得到了具有更多交互对象的序列. 接下来,需要从序列中捕获用户偏好. 受可变形内核[27]的启发,本文使用的偏移量根据更多项目综合生成内核,并基于动态卷积网络[5,28-29]提高卷积的灵活性. 结构如图2所示.
具体来说,将具有更多交互的序列映射到嵌入向量$ {\boldsymbol{h}}_{1},{\boldsymbol{h}}_{2},… ,{\boldsymbol{h}}_{n} $,$ {\boldsymbol{h}}_{i}\in {\mathbb{R}}^{d} $,n是序列的长度. 为了使卷积滤波器更灵活,根据当前输入的特征得到卷积的偏差,卷积核和原始的卷积域通过下列式(6)(7)计算得到.
$ \Delta p={linear}_{1}\left({\boldsymbol{E}}_{l}^{i}\right), $
$ {\boldsymbol{W}}_{\text{ori}}={linear}_{2}\left({\boldsymbol{E}}_{l}^{i}\right), $
其中,$ linear\left(*\right) $是线性层,$ \mathrm{\Delta }p\in {\mathbb{R}}^{1\times 2k} $是由当前输入映射而成的一系列卷积核偏差,$ {\boldsymbol{W}}_{\text{ori}}\in {\mathbb{R}}^{l\times l} $是用来生成最终卷积核的原始卷积空间. 然后使用离散内核位置上的偏移量来使固定内核更加灵活并得到最终的自适应卷积核:
$ {\boldsymbol{w}}_{i}=\mathcal G\left({\boldsymbol{W}}_{\text{ori}},\left({p}_{0}+\Delta p\right)\right). $
$ \mathcal{G}(\alpha ,\beta ) $是根据$ \beta $从原始卷积空间获取最终卷积核的过程. $ {\boldsymbol{w}}_{i}\in {\mathbb{R}}^{(n-1)\times k} $是最终的卷积核. 整个卷积的过程为:
$ DepthwiseConv\text{}\left(\boldsymbol{E},{\boldsymbol{w}}_{i},i\right)={\boldsymbol{w}}_{i}\cdot {\boldsymbol{E}}^{i-k+1:i}\text{,} $
其中,$ DepthwiseConv $是一种特殊的卷积方式,将相同卷积核应用于多个通道. 如果输出的嵌入大小(或通道数)也是d,那么标准卷积操作将需要d2l个参数,其中l为固定内核数. 为了减少参数数量,本文使用深度卷积,其中,l独立应用于每个d通道,需要dl个参数. 进一步通过对所有通道使用相同的内核,这可以进一步减少到l个参数. 参数量如此巨大的减少使得实现动态卷积变得可行,因而会产生大量计算资源.
值得注意的是为了提高训练效率和显式建模内部序列特征,本文采用Yuan等人[8]提出的简单生成结构. 生成结构一次建模所有单个项目的分布,而不仅仅是最后一个. 在序列中,具体来讲,本文使用$ \left\{{v}_{1}^{u},{v}_{2}^{u},… ,{v}_{n-1}^{u}\right\} $来预测$ \left\{{v}_{2}^{u},{v}_{3}^{u},… ,{v}_{n-2}^{u}\right\} $,而不是$ \{{v}_{1}^{u},{v}_{2}^{u},… , {v}_{n-1}^{u}\} $来预测{$ {v}_{n} $}. 由于输出与输入具有相同的形状,该结构还允许直接堆叠以扩大卷积域.
在得到卷积层的输出后,需要计算预测的物品表示和真实点击对象之间的得分. 损失函数公式[9]为:
$ \begin{split}{L}_{\mathrm{l}\mathrm{o}\mathrm{s}\mathrm{s}}=\;&\sum _{n=0}^{N}\sum _{l=2}^{L}\Bigg[\mathrm{log}\left(\sigma \left(\mathrm{cos}\left({\boldsymbol{o}}_{l}^{n},{\boldsymbol{e}}_{l}^{n}\right)\right)\right)+\\&\sum _{\bar{\boldsymbol{e}}\in {\bar{E}}^{u}}\mathrm{log}\left(1-\sigma \left(\mathrm{cos}\left({\boldsymbol{o}}_{l}^{n},\bar{\boldsymbol{e}}\right)\right)\right)\Bigg] , \end{split}$
其中N是序列的数量,L是序列的长度,$ {\boldsymbol e}_{l}^{n} $是序列中用户最终真实点击对象. $ \sigma \left(*\right) $是sigmoid函数,$ {\bar{E}}^{u} $是负例.
本文在4个公共数据集上进行模型评估,这些数据集都广泛用于序列推荐. 这些数据集的基本统计信息如表1所示. 本文训练采用前n个交互对象预测第n+1个对象,如图3所示. 前2个数据集是ml-1m和ml-20m,来自MovieLens的电影评级数据,其主要区别在于数据量. 前者包含了6038个用户的959527个交互行为,用户点击对象的数量是3706,数据的稀疏度是0.9571,后者的用户数量和交互行为明显更多一些,包含23212个用户数、1485966个点击行为以及13070个点击对象. 音乐(music)数据集是从音乐流媒体平台Last.FM收集的,包含超过10亿个音乐收听事件,本文只使用了其中一部分,用户数是102652,商品数是5735,产生的交互数量是6717493. 每个记录都包含1个时间戳,图书(book)数据集来自亚马逊,此数据集是稀疏度最高的数据集,包含的用户数量是36475,商品数是5999,产生的交互行为数量是442433.
本文考虑5种不同结构的对比模型:经典马尔可夫链、RNN、CNN、GNN和元学习. 除了 DynamicRec 和MetaTL,比较的所有模型的代码都来自RecBole平台[30-31]. 对于DynamicRec和MetaTL,使用作者发布的开源代码. 对比模型如表2所示.
1)BPR[32]. 不同于之前方法大量地使用用户显示的反馈,BPR利用用户的隐式反馈结合贝叶斯方法进行预测.
2)FPMC[1]. 该方法使用个性化马尔可夫链对序列下一行为进行推荐.
3)GRU4REC[4]. 该模型使用RNN对序列的转变模型进行建模,从而进行最后的预测.
4)Caser[7]. Caser将序列中点击对象的表示视为“image”,然后使用卷积核去捕捉序列的转换模式.
5)NextItNet[8]. 该方法使用1维的空洞卷积获取用户的偏好.
6)DynamicRec[9]. 该模型改进了普通的卷积,使用动态卷积捕捉序列行为.
7)SR-GNN[12]. 通过将序列化的问题转化为图的问题,对所有的会话序列通过有向图进行建模,使用信息聚合算法获得对象的表示向量.
为了更好地展示本文提出的GISDCN模型的特点,列出了GISDCN和其他7种模型之间的特性对比,以更好地比较它们之间的异同,如表2所示.
为了更好地评估模型的表现性能,本文采用3种主流的评价指标去测试模型的表现:平均倒数排名(MRR)、归一化折损累计增益(NDCG)、命中率(HR).MRR指标衡量的是检索到的item是否位于用户可见度较高的位置,强调位置关系和顺序性. 其公式为
$ MRR=\frac{1}{N}\sum _{i=1}^{N}\frac{1}{{p}_{i}}, $
其中,N表示推荐次数,$ {p}_{i} $表示用户真实访问的item在推荐列表中的位置,如果其没在推荐序列中,则$ p_i $为无穷大,$ 1/{p}_{i}=0 $.
NDCG是对不同用户的推荐列表的评估分数归一化,公式为
$ NDC{G}_{\rm u}@k=\frac{DC{G}_{\rm u}@k}{IDC{G}_{\rm u}}, $
其中IDCG为理想情况下最大的DCG. DCG在累计增益(CG)的基础上引入了位置影响因素,推荐结果的相关性越大,DCG越大. 相关性越好的推荐,推荐效果越好,DCG越大,计算公式如下.
$ DC{G}_{k}=\sum _{i=1}^{k}\frac{{2}^{{rel}_{i}}-1}{\mathrm{l}\mathrm{b}\left(i+1\right)}\text{,} $
其中,${rel}_{i} $用于量化推荐列表中第i个中项目与用户兴趣的匹配程序,相关性越高,取值越大.
HR反映的是在推荐序列中是否包含了用户真正点击的项目(item),公式为
$ HR=\frac{1}{N}\sum _{i=1}^{N}hit\left(i\right), $
其中, N表示推荐次数,函数$ hit\left(\mathrm{*}\right) $表示是否命中,即用户选择的item是否在推荐序列中,存在则$hit\left(\mathrm{*}\right)=1 $,反之$hit\left(\mathrm{*}\right)=0 $.
对于所有数据集,将交互记录按用户id划分,并按时间戳升序排序. 由于music数据集和book数据集中的数据量太大,通过合并重复点击的项目来过滤数据,并选择最频繁的6000个对象. 本文引入冷启动用户场景,保留了少于5次交互的部分序列. 对于没有短序列的电影数据集,通过从长序列中随机删除项目来生成序列. 控制不同比例的冷启动用户,以评估模型的效果. 然后,随机选择70%的序列作为训练集,10%的序列作为验证集,以找到模型的最佳超参数设置. 最后使用剩余20%的数据来评估模型的性能.
将GISDCN和其他模型的所有参数调整到最佳水平进行比较. 对于GISDCN,为所有数据集设置GNN嵌入大小d =200. ml-1m和book的最大长度设置为10;对于ml-20m和music,将最大长度设置为40,学习率设置为0. 001. 对于基于图的插值模块,隐藏层大小设置为100,卷积核大小设置为5,范围大小设置为7.
所有模型的总体对比结果如表3所示. 相比SR-GNN,GISDCN具有更好的性能. 主要原因分析如下:首先,SR-GNN没有很好地利用序列中的位置信息. 其次,它对这些冷启动用户序列没有任何作用. 与DynamicRec相比,由于实验中将项目序列重新构建为全局图以考虑长序列中的信息,因此GISDCN模型在处理这些冷启动用户时可以更好地利用长程依赖关系. 同时,从结果可以观察到GISDCN在ml-1m中取得的提升有限. 一个可能的原因是ml-1m的数据量很少,造成其无法完全支持训练模型.
在本节中,深入探讨了基于图的插值模块与基于可变形卷积网络的模块对推荐系统性能的影响,并在book和ml-1m数据集上进行了系统的实验与分析,以期更全面地理解这些模块的工作机制及其潜在优势.
为了证明基于图的插值模块的有效性,本文将模型的性能与不同的插值方法进行了比较,结果如图4所示,将冷启动用户的比例设置为10%. 比较的第1种方法是本文模型,它在插值过程中没有顺序限制.
第2种方法通过GNN使用核心节点的原始表示来选择候选项目. 第3种方法是本文所提出的完整模型. 可以观察到,在没有顺序限制的情况下,模型的性能甚至会降低. 随着核心节点表示的增强,模型可以增加插值的积极作用,进一步提高模型的性能.
本文比较了使用不同卷积方法的效果,证明本文基于可变形卷积网络的推荐系统的有效性. 为了防止不必要的干扰,为所有方法设置相同的冷启动用户比例(4%). 本文选择了3种方法进行比较,第1种对比方法使用动态滤波器[9,29-30],第2种对比方法是本文使用可变形卷积网络的模型,结果如表4所示. 具有动态滤波器的模型优于一般固定滤波器的模型,这证明动态滤波器在从序列中捕获复杂转换模式方面的有效性. 可变形卷积网络实现了最佳性能,综合考虑多个项目可以更好地生成鲁棒滤波器并提高序列动态的可提取性. 同时,可以发现模型在处理稀疏数据集时性能优势更加明显.
本文还对多个参数进行超参数敏感性实验,以便更好地了解它们对于模型的影响,包括GNN中的嵌入维度、CNN中商品对象的嵌入维度、卷积核大小和冷启动用户占比.
图5 (a)中GNN模块随着嵌入维度的增加,NDCG和MRR都得到了改善. 原因可能是GNN可以更好地聚合具有更高维度的项目特征信息,从而产生更合理的项目插值. 但是当隐藏大小达到一定值时,改善并不明显. 为了降低参数数量和训练时间,本文采用100作为最终隐藏大小.
本文还探讨了CNN中嵌入维度对模型性能的影响. 图5 (b)展示了在嵌入维度从50到300变化过程中模型性能的变化趋势. 从直观上看,增加嵌入维度有助于增强项目特征,从而提高模型的性能. 值得注意的是,较高的嵌入维度并不总是对应着更优的模型表现,可能是由于过拟合问题所致.
另外一方面,卷积核大小决定了可变形动态卷积核的位置可以移动的最大范围. 卷积核太小会使感受野变窄,而卷积核太大容易被噪声误导. 将卷积核从6变化为9,以研究两者之间的平衡. 在图6(a)中,随着卷积核大小的增加,模型的性能得到了提高. 但是当卷积核大小从7变化到9时,它的表现会变差,因为在不同冷启动场景下,GISDCN在所有实验条件下均优于其他对照模型,验证了其在缓解冷启动问题方面的有效性如图6(b)所示. 当冷启动用户比例设定为4%时,各方法的性能表现最佳. 值得注意的是,随着冷启动用户比例的逐渐增加,所有模型的性能均呈现下降趋势. 然而,与其他模型性能的急剧下降相比,GISDCN的性能下降幅度相对较小. 这一结果进一步证明了该模型在应对更为复杂的冷启动场景时的显著鲁棒性,展现出其在实际应用中的潜在优势.
本节对模型的有效性进行可视化分析. 在基于全局图的插值模块中,我们尝试利用所有用户的核心兴趣,对这些短序列进行补全. 本文使用 GNN找到核心兴趣节点,并将合适的物品与候选节点区分开来,以适应用户在短序列中可能的偏好. 通过从长序列中随机删除项目来生成短序列,以测试模型的效果. 图7展示了序列补全模块是如何工作的、使用的电影数据集,虚线框内的电影是从用户的原始点击序列保留下来的,方式是随机删除其他点击对象. 例如当用户最终真实的点击目标是科幻电影《超人》时,如果仅使用补全之前3部保留下来的电影进行预测,模型会得到惊悚类型电影预测,而在基于全局图信息进行补全之后模型则能够准确预测出用户最可能点击的是《超人》系列科幻电影. 这充分证明了本文模型中基于图信息插值补全短序列对于提升模型预测效果的有效性.
本文提出了一种创新性的方法,以缓解推荐系统中的冷启动问题,并结合可变形卷积滤波器以提取复杂的序列依赖关系. 研究首先将序列重构为全局图,通过GNN选择与核心兴趣节点相邻的插值项目,从而有效弥补短序列交互数据的不足. 为了捕捉动态转换模式,引入偏移量生成更具鲁棒性的滤波器,以增强卷积操作的灵活性. 实验结果基于4个真实数据集,验证了本文方法的有效性,表明本文方法在性能上优于大多数现有的先进方法.
作者贡献声明:刘昕悦提出了算法思路和实验方案并撰写论文;尹海莲、臧亚磊负责数据建模和实验;吴文隆负责实验设计与数据分析;卓俊男、徐凤如和陈吕莹负责数据采集、实验研究及结果分析;马维华提出指导意见并修改论文;李博涵提出研究构想和论文的关键修改,统筹指导整个研究过程.

参考文献 引证文献
排序方式:
1
Rendle S, Freudenthaler C, Schmidt-Thieme L. Factorizing personalized Markov chains for next-basket recommendation[C]//Proc of the 19th Int Conf on World Wide Web. New York: ACM, 2010: 811−820
2
Beutel A, Faloutsos C. User behavior modeling and fraud detection[J]. IEEE Intelligent Systems, 2016, 31(2): 84−86
3
Donkers T, Loepp B, Ziegler J. Sequential user-based recurrent neural network recommendations[C]//Proc of the 11th ACM Conf on Recommender Systems. New York: ACM, 2017: 152−160
4
Hidasi B, Karatzoglou A. Recurrent neural networks with top-k gains for session-based recommendations[C]//Proc of the 27th ACM Int Conf on Information and Knowledge Management. New York: ACM, 2018: 843−852
5
Hidasi B, Karatzoglou A, Baltrunas L, et al. Session-based recommendations with recurrent neural networks[J]. arXiv preprint, arXiv: 1511.06939, 2015
6
卢晓凯, 封军, 韩永强, 等. GraphMLP-Mixer: 基于图-多层感知机架构的高效多行为序列推荐方法[J]. 计算机研究与发展, 2024, 61(8): 1917−1929
Lu Xiaokai, Feng Jun, Han Yongqiang, et al. GraphMLP-Mixer: A graph-MLP architecture for efficient multi-behavior sequential recommendation method[J]. Journal of Computer Research and Development, 2024, 61(8): 1917−1929 (in Chinese)
7
Tang Jiaxi, Wang Ke. Personalized top-n sequential recommendation via convolutional sequence embedding[C]//Proc of the 11th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2018: 565−573
8
Yuan Fajie, Karatzoglou A, Arapakis I, et al. A simple convolutional generative network for next item recommendation[C]//Proc of the 12th ACM Int Conf on Web Search and Data Mining. New York: ACM, 2019: 582−590
9
Tanjim M M. DynamicRec: A dynamic convolutional network for next item recommendation[C]//Proc of the 29th ACM Int Conf on Information and Knowledge Management (CIKM−2020). New York: ACM, 2020: 2237−2240
10
Wang Chengkang, Julian J. Self-attentive sequential recommendation[C]//Proc of the 18th Int Conf on Data Mining. Piscataway, NJ: IEEE, 2018: 197–206
11
Li Jiacheng, Wang Yujie, McAuley J. Time interval aware self-attention for sequential recommendation[C]//Proc of the 13th Int Conf on Web Search and Data Mining. New York: ACM, 2020: 322−330
12
Wu Shu, Tang Yuyuan, Zhu Yanqiao, et al. Session-based recommendation with graph neural networks[C]//Proc of the 33rd AAAI Conf on Artificial Intelligence. Piscataway, NJ: IEEE, 2019: 346−353
13
Chang Jianxin, Gao Chen, Zheng Yu, et al. Sequential recommendation with graph neural networks[C]//Proc of the 44th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2021: 378−387
14
Zheng Yujia, Liu Siyi, Li Zekun, et al. DGTN: Dual-channel graph transition network for session-based recommendation[C]//Proc of the 20th Int Conf on Data Mining Workshops. Piscataway, NJ: IEEE, 2020: 236−242
15
He Hulingxiao, He Xiangteng, Peng Yuxin, et al. Firzen: Firing strict cold-start items with frozen heterogeneous and homogeneous graphs for recommendation[C]//Proc of IEEE 40th Int Conf on Data Engineering. Piscataway, NJ: IEEE, 2024: 4657−4670
16
Lin Xinyu, Wang Wenjie, Zhao jujia, et al. Temporally and distributionally robust optimization for cold-start recommendation[C]//Proc of the AAAI Conf on Artificial Intelligence. Palo Alto, CA: AAAI, 2024, 38(8), 8750−8758
17
Wang Yaqing, Piao Hongming, Dong Daxiang, et al. Warming up cold-start CTR prediction by learning item-specific feature interactions[C]//Proc of the 30th ACM SIGKDD Conf on Knowledge Discovery and Data Mining. New York: ACM, 2024: 3233−3244
18
Wang Jianling, Ding Kaize, Caverlee J. Sequential recommendation for cold-start users with meta transitional learning[C]//Proc of the 44th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2021: 1783−1787
19
Xu X, Dong H, Qi L, et al. CMCLR: Cross-modal contrastive learning for user cold-start sequential recommendation[C]//Proc of the 47th Int ACM SIGIR Conf on Research and Development in Information Retrieval. New York: ACM, 2024: 1589−1598
20
Hu Guangneng, Zhang Yu, Yang Qiang. Collaborative cross networks for cross-domain recommendation[C]//Proc of the 27th ACM Int Conf on Information and Knowledge Management. New York: ACM, 2018: 667−676
21
Fan Wenqi, Ma Yao, Qing Li, et al. Graph neural networks for social recommendation[C]//Proc of the 26th World Wide Web Conf. New York: ACM, 2019: 417–426
22
Xu Chengfeng, Zhao Pengpeng, Liu Yanchi, et al. Graph contextualized self-attention network for session-based recommendation [C]//Proc of the 28th Int Joint Conf on Artificial Intelligence. New York: ACM, 2019: 3940–3946
23
Yang Y, Huang C, Xia L, et al. Knowledge graph self-supervised rationalization for recommendation[C]//Proc of the 29th ACM SIGKDD Conf on Knowledge Discovery and Data Mining. New York: ACM, 2023: 3046−3056
24
Cui Zeyu, Li Zekun, Wu Shu, et al. Dressing as a whole: Outfit compatibility learning based on node-wise graph neural networks[C]//Proc of the 26th World Wide Web Conf. New York: ACM, 2019: 307–317
25
Zhang Y, Zhang Y, Zhao Y, et al. Dual Variational graph reconstruction learning for social recommendation[J]. IEEE Transactions on Knowledge and Data Engineering, 2024
26
Khan R H, Miah J, Arafat S M Y, et al. Improving traffic density forecasting in intelligent transportation systems using gated graph neural networks[C]//Proc of the 15th Int Conf on Innovations in Information Technology (IIT). Piscataway, NJ: IEEE, 2023: 104−109
27
Gao Hang, Zhu Xizhou, Lin S, et al. Deformable kernels: Adapting effective receptive fields for object deformation[J]. arXiv preprint, arXiv: 1910.02940, 2019
28
Zhang G, Hu Z, Wen G, et al. Dynamic graph convolutional networks by semi-supervised contrastive learning[J]. Pattern Recognition, 2023, 139: 109486
29
Liu Yi, Li Bohan, Zang Yalei, et al. A knowledge-aware recommender with attention-enhanced dynamic convolutiona network[C]//Proc of the 30th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2021: 1079−1088
30
Zhao Wayne, Mu Shanlei, Hou Yupeng, et al. RecBole: Towards a unified, comprehensive and efficient framework for recommendation algo-rithms[C]//Proc of the 30th ACM Int Conf on Information & Knowledge Management. New York: ACM, 2021: 4653−4664
31
曾义夫, 牟其林, 周乐, 等. 基于图表示学习的会话感知推荐模型[J]. 计算机研究与发展, 2020, 57(3): 590−603
Zeng Yifu, Mu Qilin, Zhou Le, et al. Graph embedding based session perception model for next click recommendation[J]. Journal of Computer Research and Development, 2020, 57(3): 590−603 (in Chinese)
32
Steffen R, Freudenthaler C, Gantner Z, et al. BPR: Bayesian personalized ranking from implicit feedback[J]. arXiv preprint, arXiv: 1205.2618, 2012
2025年第62卷第10期
116
52
0
引用本文
BibTeX
文章信息
doi: 10.7544/issn1000-1239.202220749
  • 接收时间:2023-04-10
  • 首发时间:2026-11-10
  • 出版时间:2025-10-01
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2023-04-10
  • 修回日期:2025-04-10
基金
作者信息
    1南京航空航天大学人工智能学院 南京 211106
    2脑机智能技术教育部重点实验室(南京航空航天大学) 南京 211106
    3软件新技术与产业化协同创新中心重点实验室 南京 211106
    4中小型无人机先进技术工业和信息化部重点实验室(南京航空航天大学) 南京 210006
    5南京航空航天大学计算机科学与技术学院 南京 211106
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202220749
分享至
全文二维码

扫描看全文

引用本文
BibTeX
本文的引用情况
2种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏