Article(id=1327012441898201188, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327012311115608215, articleNumber=null, orderNo=null, doi=10.7544/issn1000-1239.202330428, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=research-article, receivedDate=1685894400000, receivedDateStr=2023-06-05, revisedDate=1707148800000, revisedDateStr=2024-02-06, acceptedDate=null, acceptedDateStr=null, onlineDate=1794314020846, onlineDateStr=2026-11-10, pubDate=1738339200000, pubDateStr=2025-02-01, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1794314020846, onlineIssueDateStr=2026-11-10, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1794314020846, creator=13701087609, updateTime=1794314020846, updator=13701087609, issue=Issue{id=1327012311115608215, tenantId=1146029695717560320, journalId=1324759914146529478, year='2025', volume='62', issue='2', pageStart='285', pageEnd='541', issueExtLink='null', onlineDate='null', pubDate='1738339200000', pubDateStr='2025-02-01', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=null, createTime=1794313989665, creator='13701087609', updateTime=1794376696965, updator='13701087609', preIssue=null, nextIssue=null, articleTotal=null, ext={EN=IssueExt(id=1327275325060395748, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327012311115608215, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1327275325060395749, tenantId=1146029695717560320, journalId=1324759914146529478, issueId=1327012311115608215, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null, downloadFileDto=null}, startPage=383, endPage=396, ext={EN=ArticleExt(id=1327012442116304997, articleId=1327012441898201188, tenantId=1146029695717560320, journalId=1324759914146529478, language=EN, title=R-FCCL: An Approach of Fuzzy-Based Concept-Cognitive Learning with Robustness for High-Dimensional Data, columnId=1325489891789910943, journalTitle=Journal of Computer Research and Development, columnName=Artificial Intelligence, runingTitle=null, highlight=null, articleAbstract=

With the rapid development of global informationization, data mining and knowledge discovery of high-dimensional data have been a hotspot in the field of artificial intelligence and data science. However, the sparse sample and redundant feature issues of high-dimensional data make it challenging to ensure the generalization and interpretability of traditional statistical models and machine learning methods. Hence, we present fuzzy-based concept-cognitive learning with robustness for the imbalance problem between high-dimensional data and weak knowledge evolution ability. The main idea is to explore the knowledge structure and cognitive learning mechanism of high-dimensional data from the concept perspective. We propose a high-dimensional data classification method based on the concept-cognitive learning mechanism in the fuzzy formal context. Furthermore, the cognitive learning process of fuzzy concepts is described from two different perspectives by the positive and negative cognitive learning operators of fuzzy three-way concepts. Finally, the fusion of fuzzy three-way concepts completes the task of concept identification and data classification. Extensive experiments performed on 12 real data sets compared with 12 state-of-the-art classification methods also verify the robustness and effectiveness of the proposed method. The considered framework can provide a convenient novel tool for high-dimensional data knowledge discovery research and fuzzy-based concept-cognitive learning.

, authors=Doudou Guo, Weihua Xu*, authorsList=Doudou Guo, Weihua Xu, authorCompany=null, correspAuthors=Weihua Xu, authorNote=

born in 1995. PhD candidate, Member of CCF. His main research interests include concept-cognitive learning, granular computing, and machine learning

born in 1979. PhD, professor, PhD supervisor. His main research interests include granular computing, cognitive computing, data mining, and machine learning

, 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=1327012443584311408, articleId=1327012441898201188, tenantId=1146029695717560320, journalId=1324759914146529478, language=CN, title=R-FCCL:一种面向高维数据的稳健模糊概念认知学习方法, columnId=1325489892104483745, journalTitle=计算机研究与发展, columnName=人工智能, runingTitle=null, highlight=null, articleAbstract=

随着全球信息化的高速发展,高维数据挖掘与知识发现成为了人工智能领域亟待破解的科学问题之一. 然而,由于高维数据中样本的稀疏性与特征的冗余性,传统统计学模型和机器学习方法的泛化性和可解释性遇到极大的挑战. 为此,针对高维数据与知识弱演化能力之间不平衡的科学问题,利用三支概念求解复杂问题的认知机理,提出了一种新的概念建模方法,即稳健模糊概念认知学习(fuzzy-based concept-cognitive learning with robustness,R-FCCL). 首先,借助概念的最大相似性原则,建立了基于R-FCCL的高维数据分类系统,并从概念的角度出发,研究了高维数据的知识结构和认知学习机理. 进一步,利用模糊三支概念的正、负算子从2个不同的角度刻画了模糊环境概念认知学习过程,进而基于概念融合的模糊三支概念完成概念辨识和数据分类. 通过在12个真实数据集与12种分类方法的实验分析,验证所提方法具有较好的鲁棒性和有效性.

, authors=郭豆豆, 徐伟华*, authorsList=郭豆豆, 徐伟华, authorCompany=null, correspAuthors=徐伟华, authorNote=

1995年生. 博士研究生. CCF会员. 主要研究方向为概念认知学习、粒计算、机器学习

1979年生. 博士,教授,博士生导师. 主要研究方向为粒计算、认知计算、数据挖掘、机器学习

, correspAuthorsNote=
徐伟华()
, copyrightStatement=版权所有 © 2025 计算机研究与发展编辑部, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=qDIKV6WkfmNeC6Dm5X5eBw==, magXml=dkRXKsGIc5f3hgALZ6FpUQ==, pdfUrl=null, pdf=n3+Gf92Qvpw4Pjy9qBH5NQ==, pdfFileSize=2258959, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=1X45tvnWKwKbhEFcuXRxsQ==, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=cZWbJlRRtEj+tEmzbsk+fQ==, mapNumber=null, fund=null)}, authors=[Author(id=1327012443919855733, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=doudou876517690@126.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012443991158903, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012443919855733, language=EN, stringName=Doudou Guo, firstName=Doudou, middleName=null, lastName=Guo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Artificial Intelligence, Southwest University, Chongqing 400715, bio={"content":"

born in 1995. PhD candidate, Member of CCF. His main research interests include concept-cognitive learning, granular computing, and machine learning

"}, bioImg=null, bioContent=

born in 1995. PhD candidate, Member of CCF. His main research interests include concept-cognitive learning, granular computing, and machine learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012444070850680, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012443919855733, language=CN, stringName=郭豆豆, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=西南大学人工智能学院 重庆 400715, bio={"img":"kxPEdRGh/RpZzMHtPQh4Cw==","content":"

1995年生. 博士研究生. CCF会员. 主要研究方向为概念认知学习、粒计算、机器学习

"}, bioImg=kxPEdRGh/RpZzMHtPQh4Cw==, bioContent=

1995年生. 博士研究生. CCF会员. 主要研究方向为概念认知学习、粒计算、机器学习

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012443827581041, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, xref=null, ext=[AuthorCompanyExt(id=1327012443835969650, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Artificial Intelligence, Southwest University, Chongqing 400715), AuthorCompanyExt(id=1327012443848552563, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西南大学人工智能学院 重庆 400715)])]), Author(id=1327012444133765242, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chxuwh@gmail.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1327012444217651324, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012444133765242, language=EN, stringName=Weihua Xu, firstName=Weihua, middleName=null, lastName=Xu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=College of Artificial Intelligence, Southwest University, Chongqing 400715, bio={"content":"

born in 1979. PhD, professor, PhD supervisor. His main research interests include granular computing, cognitive computing, data mining, and machine learning

"}, bioImg=null, bioContent=

born in 1979. PhD, professor, PhD supervisor. His main research interests include granular computing, cognitive computing, data mining, and machine learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012444288954493, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012444133765242, language=CN, stringName=徐伟华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=西南大学人工智能学院 重庆 400715, bio={"img":"SdG9RUMTN6GtS4OA/Lu6xA==","content":"

1979年生. 博士,教授,博士生导师. 主要研究方向为粒计算、认知计算、数据挖掘、机器学习

"}, bioImg=SdG9RUMTN6GtS4OA/Lu6xA==, bioContent=

1979年生. 博士,教授,博士生导师. 主要研究方向为粒计算、认知计算、数据挖掘、机器学习

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012443827581041, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, xref=null, ext=[AuthorCompanyExt(id=1327012443835969650, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Artificial Intelligence, Southwest University, Chongqing 400715), AuthorCompanyExt(id=1327012443848552563, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西南大学人工智能学院 重庆 400715)])])], keywords=[Keyword(id=1327012444389617790, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, orderNo=1, keyword=formal concept analysis), Keyword(id=1327012444452532351, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, orderNo=2, keyword=concept-cognitive learning), Keyword(id=1327012444515446912, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, orderNo=3, keyword=granular computing), Keyword(id=1327012444586750081, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, orderNo=4, keyword=high-dimensional data), Keyword(id=1327012444670636162, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, orderNo=5, keyword=three-way decision), Keyword(id=1327012444733550723, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, orderNo=1, keyword=形式概念分析), Keyword(id=1327012444796465284, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, orderNo=2, keyword=概念认知学习), Keyword(id=1327012444863574149, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, orderNo=3, keyword=粒计算), Keyword(id=1327012444926488710, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, orderNo=4, keyword=高维数据), Keyword(id=1327012444997791879, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, orderNo=5, keyword=三支决策)], refs=[Reference(id=1327012446537101464, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, 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=于洪,何德牛,王国胤,等. 大数据智能决策[J]. 自动化学报,2020,46(5):878−896, articleTitle=null, refAbstract=null), Reference(id=1327012446604210329, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=1, rfOrder=1, authorNames=null, journalName=null, refType=null, unstructuredReference=Yu Hong, He Deniu, Wang Guoyin, et al. Big data for intelligent decision making[J]. Acta Automatica Sinica, 2020, 46(5): 878−896 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012446654541978, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=2, authorNames=null, journalName=null, refType=null, unstructuredReference=于晓,刘慧,吴彦,等. 基于本质自表示的多视角子空间聚类[J]. 中国科学:信息科学,2021,51(10):1625−1639, articleTitle=null, refAbstract=null), Reference(id=1327012446713262235, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=2, rfOrder=3, authorNames=null, journalName=null, refType=null, unstructuredReference=Yu Xiao, Liu Hui, Wu Yan, et al. Intrinsic self-representation for multi-view subspace clustering[J]. SCIENTIA SINICA Informations, 2021, 51(10): 1625−1639 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012446780371100, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=3, rfOrder=4, authorNames=null, journalName=null, refType=null, unstructuredReference=Cózar J, Fernández A, Herrera F, et al. A metahierarchical rule decision system to design robust fuzzy classifiers based on data complexity[J]. IEEE Transactions on Fuzzy Systems, 2019, 27(4): 701−715, articleTitle=null, refAbstract=null), Reference(id=1327012446847479965, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=4, rfOrder=5, authorNames=null, journalName=null, refType=null, unstructuredReference=Gómez D, Rojas A. An empirical overview of the no free lunch theorem and its effect on real-world machine learning classification[J]. Neural Computing, 2016, 28(1): 216−228, articleTitle=null, refAbstract=null), Reference(id=1327012446910394526, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=5, rfOrder=6, authorNames=null, journalName=null, refType=null, unstructuredReference=Lake B M, Salakhutdinov R, Tenenbaum J B, et al. Human-level concept learning through probabilistic program induction[J]. Science, 2015, 350(6266): 1332−1338, articleTitle=null, refAbstract=null), Reference(id=1327012446977503391, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=7, authorNames=null, journalName=null, refType=null, unstructuredReference=于剑. 机器学习:从公理到算法[M]. 北京:清华大学出版社,2017, articleTitle=null, refAbstract=null), Reference(id=1327012447040417952, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=6, rfOrder=8, authorNames=null, journalName=null, refType=null, unstructuredReference=Yu Jian. Machine Learning from Axioms to Algorithms[M]. Beijing: Tsinghua University Press, 2017 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012447115915425, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=9, authorNames=null, journalName=null, refType=null, unstructuredReference=李金海,米允龙,刘文奇. 概念的渐进式认知理论与方法[J]. 计算机学报,2019,42(10):2233−2250, articleTitle=null, refAbstract=null), Reference(id=1327012447195607202, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=7, rfOrder=10, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jinhai, Mi Yunlong, Liu Wenqi. Incremental cognitive of concept: Theories and methods[J]. Chinese Journal of Computers, 2019, 42(10): 2233−2250 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012447266910371, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=11, authorNames=null, journalName=null, refType=null, unstructuredReference=李金海,李玉斐,米允龙,等. 多粒度形式概念分析的介粒度标记方法[J]. 计算机研究与发展,2020,57(2):447−458, articleTitle=null, refAbstract=null), Reference(id=1327012447346602148, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=8, rfOrder=12, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jinhai, Li Yufei, Mi Yunlong, et al. Meso-granularity labeled method for multi-granularity formal concept analysis[J]. Journal of Computer Research and Development, 2020, 57(2): 447−458 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012447417905317, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=13, authorNames=null, journalName=null, refType=null, unstructuredReference=张文修,徐伟华. 基于粒计算的认知模型[J]. 工程数学学报,2007,24(6):957−971, articleTitle=null, refAbstract=null), Reference(id=1327012447497597094, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=9, rfOrder=14, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Wenxiu, Xu Weihua. Cognitive model based on granular computing[J]. Chinese Journal of Engineering Mathematics, 2007, 24(6): 957−971 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012447556317351, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=10, rfOrder=15, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jinhai, Mei Changlin, Xu Weihua, et al. Concept learning via granular computing: A cognitive viewpoint[J]. Information Sciences, 2015, 298: 447−467, articleTitle=null, refAbstract=null), Reference(id=1327012447640203432, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=16, authorNames=null, journalName=null, refType=null, unstructuredReference=王霞,江山,李俊余,等. 三元概念的一种构造方法[J]. 计算机研究与发展,2019,56(4):844−853, articleTitle=null, refAbstract=null), Reference(id=1327012447698923689, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=11, rfOrder=17, authorNames=null, journalName=null, refType=null, unstructuredReference=Wang Xia, Jiang Shan, Li Junyu, et al. A construction method of triadic concepts[J]. Journal of Computer Research and Development, 2019, 56(4): 844−853 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012447774421162, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=12, rfOrder=18, authorNames=null, journalName=null, refType=null, unstructuredReference=Yao Yiyu. Interval sets and three-way concept analysis in incomplete contexts[J]. International Journal of Machine Learning and Cybernetics, 2017, 8: 3−20, articleTitle=null, refAbstract=null), Reference(id=1327012447883473067, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=13, rfOrder=19, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Weihua, Guo Doudou, Mi Jusheng, et al. Two-way concept-cognitive learning via concept movement viewpoint[J]. IEEE Transactions on Neural Networks and Learning Systems. 2023, 34(10): 6798−6812, articleTitle=null, refAbstract=null), Reference(id=1327012447971553452, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=14, rfOrder=20, authorNames=null, journalName=null, refType=null, unstructuredReference=Mi Yunlong, Shi Yong, Li Jinhai, et al. Fuzzy-based concept learning method: Exploiting data with fuzzy conceptual clustering[J]. IEEE Transactions on Cybernetics, 2021, 51(2): 809−821, articleTitle=null, refAbstract=null), Reference(id=1327012448034468013, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=15, rfOrder=21, authorNames=null, journalName=null, refType=null, unstructuredReference=Shi Yong, Mi Yunlong, Li Jinhai, et al. Concurrent concept-cognitive learning model for classification[J]. Information Sciences, 2019, 496: 65−81, articleTitle=null, refAbstract=null), Reference(id=1327012448114159791, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=16, rfOrder=22, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Xiaoyan, Guo Doudou, Xu Weihua. Two-way concept-cognitive learning with multi-source fuzzy context[J]. Cognitive Computation, 2023, 15(5): 1526−1548, articleTitle=null, refAbstract=null), Reference(id=1327012448185462960, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=17, rfOrder=23, authorNames=null, journalName=null, refType=null, unstructuredReference=Yan Enliang, Yu Cunguo, Lu Liming, et al. Incremental concept cognitive learning based on three-way partial order structure[J]. Knowledge-Based Systems, 2021, 220: 68−98, articleTitle=null, refAbstract=null), Reference(id=1327012448256766129, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=18, rfOrder=24, authorNames=null, journalName=null, refType=null, unstructuredReference=Hao Fei, Pang Guangyao, Pei Zheng, et al. Virtual machines scheduling in mobile edge computing: A formal concept analysis approach[J]. IEEE Transactions on Sustainable Computing, 2019, 5(3): 319−328, articleTitle=null, refAbstract=null), Reference(id=1327012448319680690, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=25, authorNames=null, journalName=null, refType=null, unstructuredReference=魏玲,曹丽,祁建军,等. 形式概念分析中的概念约简与概念特征[J]. 中国科学:信息科学,2020,50(12):1817−1833, articleTitle=null, refAbstract=null), Reference(id=1327012448395178163, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=19, rfOrder=26, authorNames=null, journalName=null, refType=null, unstructuredReference=Wei Ling, Cao Li, Qi Jianjun, et al. Concept reduction and concept characteristics in formal concept analysis[J]. SCIENTIA SINICA Informations, 2020, 50(12): 1817−1833 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012448470675636, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=27, authorNames=null, journalName=null, refType=null, unstructuredReference=徐伟华,杨蕾,张晓燕. 模糊三支形式概念分析与概念认知学习[J]. 西北大学学报:自然科学版,2020,50(4):516−528, articleTitle=null, refAbstract=null), Reference(id=1327012448617476277, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=20, rfOrder=28, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Weihua, Yang Lei, Zhang Xiaoyan. Fuzzy three-way formal concept analysis and concept-cognitive learning[J]. Journal of Northwest University: Natural Science Edition, 2020, 50(4): 516−528 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012448713945270, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=21, rfOrder=29, authorNames=null, journalName=null, refType=null, unstructuredReference=Qian Ting, Wei Ling, Qi Jianjun. Constructing three-way concept lattices based on apposition and subposition of formal contexts[J]. Knowledge-Based Systems, 2017, 116(15): 39−48, articleTitle=null, refAbstract=null), Reference(id=1327012448802025655, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=22, rfOrder=30, authorNames=null, journalName=null, refType=null, unstructuredReference=Guo Doudou, Xu Weihua. Fuzzy-based concept-cognitive learning: An investigation of novel approach to tumor diagnosis analysis[J]. Information Sciences, 2023, 639: 118998, articleTitle=null, refAbstract=null), Reference(id=1327012448885911736, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=31, authorNames=null, journalName=null, refType=null, unstructuredReference=马娜,范敏,李金海. 复杂网络下的概念认知学习[J]. 南京大学学报:自然科学,2019,55(4):609−623, articleTitle=null, refAbstract=null), Reference(id=1327012448944631993, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=23, rfOrder=32, authorNames=null, journalName=null, refType=null, unstructuredReference=Ma Na, Fan Min, Li Jinhai. Concept-cognitive learning under complex network[J]. Journal of Nanjing University: Natural Science, 2019, 55(4): 609−623 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012449032712378, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=24, rfOrder=33, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Weihua, Guo Doudou, Qian Yuhua, et al. Two-way concept-cognitive learning method: A fuzzy-based progressive learning[J]. IEEE Transactions on Fuzzy Systems, 2023, 31(6): 1885−1899, articleTitle=null, refAbstract=null), Reference(id=1327012449112404155, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=25, rfOrder=34, authorNames=null, journalName=null, refType=null, unstructuredReference=Fan Bingjiao, Tsang E C C, Xu Weihua, et al. Attribute-oriented cognitive concept learning strategy: A multi-level method[J]. International Journal of Machine Learning and Cybernetics, 2019, 10(9): 2421−2437, articleTitle=null, refAbstract=null), Reference(id=1327012449187901628, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=26, rfOrder=35, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jinhai, Huang Chenchen, Qi Jianjun, et al. Three-way cognitive concept learning via multi-granularity[J]. Information Sciences, 2017, 378: 244−263, articleTitle=null, refAbstract=null), Reference(id=1327012449305342141, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=27, rfOrder=36, authorNames=null, journalName=null, refType=null, unstructuredReference=Guo Doudou, Xu Weihua, Qian Yuhua, et al. Fuzzy-granular concept-cognitive learning via three-way decision: Performance evaluation on dynamic knowledge discovery[J/OL]. IEEE Transactions on Fuzzy Systems, 2023[2023-10-19]. https://ieeexplore.ieee.org/document/10288039, articleTitle=null, refAbstract=null), Reference(id=1327012449406005438, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=28, rfOrder=37, authorNames=null, journalName=null, refType=null, unstructuredReference=Guo Doudou, Xu Weihua, Qian Yuhua, et al. M-FCCL: Memory-based concept-cognitive learning for dynamic fuzzy data classification and knowledge fusion[J]. Information Fusion, 2023, 100: 101962, articleTitle=null, refAbstract=null), Reference(id=1327012449519251647, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=29, rfOrder=38, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhang Tao, Rong Mei, Shan Haoran, et al. Causal asymmetry analysis in the view of concept-cognitive learning by incremental concept tree[J]. Cognitive Computation, 2021, 13(5): 1274−1286, articleTitle=null, refAbstract=null), Reference(id=1327012449577971905, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=30, rfOrder=39, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Weihua, Li Wentao. Granular computing approach to two-way learning based on formal concept analysis in fuzzy dataset[J]. IEEE Transactions on Cybernetics, 2016, 46(2): 66−379, articleTitle=null, refAbstract=null), Reference(id=1327012449636692162, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=31, rfOrder=40, authorNames=null, journalName=null, refType=null, unstructuredReference=Ganter B. Formal Concept Analysis: Mathematical Foundations[M]. Berlin: Springer, 1999, articleTitle=null, refAbstract=null), Reference(id=1327012449703801027, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=41, authorNames=null, journalName=null, refType=null, unstructuredReference=徐伟华,李金海,魏玲,等. 形式概念分析理论与应用[M]. 北京:科学出版社,2016, articleTitle=null, refAbstract=null), Reference(id=1327012449770909892, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=32, rfOrder=42, authorNames=null, journalName=null, refType=null, unstructuredReference=Xu Weihua, Li Jinhai, Wei Ling, et al. Formal Concept Analysis: Theory and Application[M]. Beijing: Science Press, 2016 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012449879961797, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=43, authorNames=null, journalName=null, refType=null, unstructuredReference=祁建军,魏玲,姚一豫. 三支概念分析与决策[M]. 北京:科学出版社,2019, articleTitle=null, refAbstract=null), Reference(id=1327012449947070662, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=33, rfOrder=44, authorNames=null, journalName=null, refType=null, unstructuredReference=Qi Jianjun, Wei Ling, Yao Yiyu. Three-Way Concept Analysis and Decision[M]. Beijing: Science Press, 2019 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012450014179527, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=34, rfOrder=45, authorNames=null, journalName=null, refType=null, unstructuredReference=Yuan Kehua, Xu Weihua, Li Wentao, et al. An incremental learning mechanism for object classification based on progressive fuzzy three-way concept[J]. Information Sciences, 2022, 584: 127−147, articleTitle=null, refAbstract=null), Reference(id=1327012450089677000, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=35, rfOrder=46, authorNames=null, journalName=null, refType=null, unstructuredReference=Li Jundong, Cheng Kewei, Wang Suhang, et al. Feature selection: A data perspective[J]. ACM Computing Surveys, 2018, 50(6): 94−139, articleTitle=null, refAbstract=null), Reference(id=1327012450177757385, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=36, rfOrder=47, authorNames=null, journalName=null, refType=null, unstructuredReference=Elhadad A, Ghareeb A, Abbas S. A blind and high-capacity data hiding of DICOM medical images based on fuzzification concepts[J]. Alexandria Engineering Journal, 2021, 66(2): 2471−2482, articleTitle=null, refAbstract=null), Reference(id=1327012450240671946, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=37, rfOrder=48, authorNames=null, journalName=null, refType=null, unstructuredReference=Kuncheva L I. An intuitionistic fuzzy k-nearest neighbors rule[J]. Notes on Intuitionistic Fuzzy Sets, 1995, 1: 56−60, articleTitle=null, refAbstract=null), Reference(id=1327012450307780811, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=38, rfOrder=49, authorNames=null, journalName=null, refType=null, unstructuredReference=Yang M S, Cheng C H. On the edited fuzzy k-nearest neighbor rule[J]. IEEE Transactions on Systems, Man, and Cybernetics: Cybernetics, 1998, 28(8): 461−466, articleTitle=null, refAbstract=null), Reference(id=1327012450370695372, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=39, rfOrder=50, authorNames=null, journalName=null, refType=null, unstructuredReference=Arif M, Akram M U, Afsar F A, et al. Pruned fuzzy k-nearest neighbor classifier for beat classification[J]. Journal of Biomedical Science and Engineering, 2010, 3(4): 380−389, articleTitle=null, refAbstract=null), Reference(id=1327012450429415629, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=40, rfOrder=51, authorNames=null, journalName=null, refType=null, unstructuredReference=Jensen R, Cornelis C. Fuzzy-rough nearest neighbour classification[C] // Proc of the 7th Rough Sets and Current Trends in Computing Conf. Berlin: Springer, 2011: 56−72, articleTitle=null, refAbstract=null), Reference(id=1327012450496524494, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=41, rfOrder=52, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhai Junhai, Li Na, Zhai Mengyao. The condensed fuzzy k-nearest neighbor rule based on sample fuzzy entropy[C] // Proc of the 24th Int Conf on Machine Learning and Cybernetics. Berlin: Springer, 2011: 282−286, articleTitle=null, refAbstract=null), Reference(id=1327012450647519439, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=42, rfOrder=53, authorNames=null, journalName=null, refType=null, unstructuredReference=周志华. 机器学习[M]. 北京:清华大学出版社,2016, articleTitle=null, refAbstract=null), Reference(id=1327012450723016912, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=42, rfOrder=54, authorNames=null, journalName=null, refType=null, unstructuredReference=Zhou Zhihua. Machine Learning[M]. Beijing: Tsinghua University Press, 2016 (in Chinese), articleTitle=null, refAbstract=null), Reference(id=1327012450785931473, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=43, rfOrder=55, authorNames=null, journalName=null, refType=null, unstructuredReference=Adnan M N, Islam M Z. Forest PA: Constructing a decision forest by penalizing attributes used in previous trees[J]. Expert Systems with Applications, 2017, 89(15): 389−403, articleTitle=null, refAbstract=null), Reference(id=1327012450848846034, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=44, rfOrder=56, authorNames=null, journalName=null, refType=null, unstructuredReference=De G M, Maurizio G. An experimental evaluation of weightless neural networks for multi-class classification[J]. Applied Soft Computing, 2018, 72: 338−354, articleTitle=null, refAbstract=null), Reference(id=1327012450907566291, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=45, rfOrder=57, authorNames=null, journalName=null, refType=null, unstructuredReference=Yates D, Islam M Z, Gao J. SPAARC: A Fast Decision Tree Algorithm[M]. Berlin: Springer, 2019, articleTitle=null, refAbstract=null), Reference(id=1327012450970480852, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=46, rfOrder=58, authorNames=null, journalName=null, refType=null, unstructuredReference=Ii A, Jmp B, Jm B, et al. PCTBagging: From inner ensembles to ensembles. A trade-off between discriminating capacity and interpretability [J]. Information Sciences, 2022, 583: 219−238, articleTitle=null, refAbstract=null), Reference(id=1327012451029201109, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, doi=null, pmid=null, pmcid=null, year=null, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=47, rfOrder=59, authorNames=null, journalName=null, refType=null, unstructuredReference=Vlachos M, Domeniconi C, Gunopulos D, et al. Non-linear dimensionality reduction techniques for classification and visualization[C]//Proc of the 8th ACM Int Conf on Knowledge Discovery and Data Mining. New York: ACM, 2002: 645−651, articleTitle=null, refAbstract=null)], funds=null, companyList=[AuthorCompany(id=1327012443827581041, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, xref=null, ext=[AuthorCompanyExt(id=1327012443835969650, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Artificial Intelligence, Southwest University, Chongqing 400715), AuthorCompanyExt(id=1327012443848552563, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西南大学人工智能学院 重庆 400715)])], figs=[ArticleFig(id=1327012445157175432, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Figure 1, caption=Comparison of classification performance, figureFileSmall=omaLLHQfcWwMTkPBzHus5w==, figureFileBig=1X45tvnWKwKbhEFcuXRxsQ==, tableContent=null), ArticleFig(id=1327012445245255817, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=图1, caption=分类性能比较, figureFileSmall=omaLLHQfcWwMTkPBzHus5w==, figureFileBig=1X45tvnWKwKbhEFcuXRxsQ==, tableContent=null), ArticleFig(id=1327012445484331146, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Figure 2, caption=Sensitivity analysis of parameter, figureFileSmall=MChEI9wJ9E0c8gpRH7uuIw==, figureFileBig=idKaQFVcgQCF9hQF5VRdOg==, tableContent=null), ArticleFig(id=1327012445559828619, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=图2, caption=参数敏感度分析, figureFileSmall=MChEI9wJ9E0c8gpRH7uuIw==, figureFileBig=idKaQFVcgQCF9hQF5VRdOg==, tableContent=null), ArticleFig(id=1327012445639520396, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Figure 3, caption=Robustness analysis, figureFileSmall=To3SuKg9LNFGTThOBIlROg==, figureFileBig=R804WYausruX21RRdbTH4g==, tableContent=null), ArticleFig(id=1327012445702434957, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=图3, caption=鲁棒性分析, figureFileSmall=To3SuKg9LNFGTThOBIlROg==, figureFileBig=R804WYausruX21RRdbTH4g==, tableContent=null), ArticleFig(id=1327012445761155214, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Table 1, caption=Example of Fuzzy Formal Context, figureFileSmall=null, figureFileBig=null, tableContent=
对象${a_1}$${a_2}$${a_3}$${a_4}$$d$
${x_1}$0.830.730.590.531
${x_2}$0.600.890.600.501
${x_3}$0.440.310.090.471
${x_4}$0.370.430.130.751
${x_5}$0.700.070.210.140
${x_6}$0.610.010.100.120
${x_7}$0.650.070.390.060
${x_8}$0.670.060.440.060
${x_9}$0.810.070.390.060
), ArticleFig(id=1327012445849235599, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=表1, caption=

模糊形式背景示例

, figureFileSmall=null, figureFileBig=null, tableContent=
对象${a_1}$${a_2}$${a_3}$${a_4}$$d$
${x_1}$0.830.730.590.531
${x_2}$0.600.890.600.501
${x_3}$0.440.310.090.471
${x_4}$0.370.430.130.751
${x_5}$0.700.070.210.140
${x_6}$0.610.010.100.120
${x_7}$0.650.070.390.060
${x_8}$0.670.060.440.060
${x_9}$0.810.070.390.060
), ArticleFig(id=1327012445920538768, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Table 2, caption=Basic Information of Experimental Datasets, figureFileSmall=null, figureFileBig=null, tableContent=
序号数据集样本维数类别连续型
1Allaml7771292是
2Cll_Sub_111111113403是
3Colon6220002否
4Glioma5044344是
5Leukemia7270702否
6Lung20333125是
7Lung_Discrete733257否
8Orlraws10P1001030410是
9Prostate_Ge10259662是
10Tox_17117157484是
11WarpAR10P130240010是
12WarpPIE10P210242010是
), ArticleFig(id=1327012445979259025, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=表2, caption=

实验数据集的基本信息

, figureFileSmall=null, figureFileBig=null, tableContent=
序号数据集样本维数类别连续型
1Allaml7771292是
2Cll_Sub_111111113403是
3Colon6220002否
4Glioma5044344是
5Leukemia7270702否
6Lung20333125是
7Lung_Discrete733257否
8Orlraws10P1001030410是
9Prostate_Ge10259662是
10Tox_17117157484是
11WarpAR10P130240010是
12WarpPIE10P210242010是
), ArticleFig(id=1327012446042173586, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Table 3, caption=Basic Information of Selected Comparative Methods in Experiment, figureFileSmall=null, figureFileBig=null, tableContent=
方法方法描述年份
IFKNN[37]Intuitionistic Fuzzy Sets k-Nearest Neighbors Classifier.1995
FENN[38]Fuzzy Edited Nearest Neighbor Classifier.1998
PFKNN[39]Pruned Fuzzy k-Nearest Neighbors Classifier.2010
FRNN-FRS[40]Fuzzy-Rough Nearest Neighbor Classifier-Fuzzy Rough Sets2011
CFKNN[41]Condensed Fuzzy K-Nearest Neighbors Classifier2011
SVM[42]Support Vector Machine Classifier2016
KNN[42]K-Nearest Neighbor Classifier2016
ForestPA[43]Forest by Penalizing Attributes Algorithm2017
WiSARD[44]WiSARD Classifier2018
SPAARC[45]Speeding up the Decision Tree Induction Process Algorithm2019
PCTBagging[46]Partially Consolidated Tree Bagging2022
ILMPFTC[34]Incremental Learning Mechanism based on Progressive Fuzzy Three-way Concept2022
), ArticleFig(id=1327012446134448275, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=表3, caption=

实验中所选对比方法的基本信息

, figureFileSmall=null, figureFileBig=null, tableContent=
方法方法描述年份
IFKNN[37]Intuitionistic Fuzzy Sets k-Nearest Neighbors Classifier.1995
FENN[38]Fuzzy Edited Nearest Neighbor Classifier.1998
PFKNN[39]Pruned Fuzzy k-Nearest Neighbors Classifier.2010
FRNN-FRS[40]Fuzzy-Rough Nearest Neighbor Classifier-Fuzzy Rough Sets2011
CFKNN[41]Condensed Fuzzy K-Nearest Neighbors Classifier2011
SVM[42]Support Vector Machine Classifier2016
KNN[42]K-Nearest Neighbor Classifier2016
ForestPA[43]Forest by Penalizing Attributes Algorithm2017
WiSARD[44]WiSARD Classifier2018
SPAARC[45]Speeding up the Decision Tree Induction Process Algorithm2019
PCTBagging[46]Partially Consolidated Tree Bagging2022
ILMPFTC[34]Incremental Learning Mechanism based on Progressive Fuzzy Three-way Concept2022
), ArticleFig(id=1327012446201557140, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Table 4, caption=Classification Performance of R-FCCL and Six Fuzzy-Based Nearest-Neighbor Classification Methods, figureFileSmall=null, figureFileBig=null, tableContent=
序号R-FCCL(本文)KNNFENNIFKNNPFKNNFRNN-FRSCFKNN
注:“±”前的数值为准确率,“±”后的数值为标准差.
11.0000±0.00000.8607±0.14300.8231±0.15540.8864±0.11480.8864±0.11480.7289±0.12230.7987±0.1412
20.8106±0.07990.6722±0.12760.5441±0.08380.5826±0.09610.5096±0.11250.6240±0.12540.5909±0.1109
30.9357±0.08330.6818±0.18030.6991±0.15340.6948±0.18350.6515±0.18030.6407±0.04240.6905±0.1942
40.8800±0.10330.6909±0.15640.8000±0.17060.8000±0.17060.8182±0.13360.3455±0.12330.7636±0.1666
51.0000±0.00000.9010±0.10560.8506±0.09210.8766±0.10890.8766±0.09020.6510±0.06110.8247±0.1057
60.9602±0.03940.9329±0.03150.9465±0.03870.9597±0.03560.9333±0.05150.8660±0.05250.9513±0.0411
70.8893±0.05880.8393±0.11750.7662±0.09940.8555±0.13090.8539±0.11940.1347±0.08140.8929±0.0911
80.9600±0.05160.9455±0.06560.8800±0.07480.9500±0.06710.7300±0.06400.6800±0.13270.9100±0.0943
90.9300±0.11600.8314±0.08430.8405±0.12830.8488±0.12360.8570±0.13140.6058±0.11800.5653±0.1443
100.9650±0.04090.8520±0.06750.6364±0.10810.8164±0.09400.7362±0.07310.7885±0.09100.6043±0.0787
110.8231±0.08920.4965±0.13670.4126±0.19200.3986±0.13850.3636±0.18290.4685±0.16550.4056±0.1576
121.0000±0.00000.9913±0.02740.9177±0.07890.9221±0.05840.7922±0.06510.9740±0.05510.6840±0.0819
), ArticleFig(id=1327012446268666005, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=表4, caption=

R-FCCL与6种基于模糊最近邻分类方法的分类性能

, figureFileSmall=null, figureFileBig=null, tableContent=
序号R-FCCL(本文)KNNFENNIFKNNPFKNNFRNN-FRSCFKNN
注:“±”前的数值为准确率,“±”后的数值为标准差.
11.0000±0.00000.8607±0.14300.8231±0.15540.8864±0.11480.8864±0.11480.7289±0.12230.7987±0.1412
20.8106±0.07990.6722±0.12760.5441±0.08380.5826±0.09610.5096±0.11250.6240±0.12540.5909±0.1109
30.9357±0.08330.6818±0.18030.6991±0.15340.6948±0.18350.6515±0.18030.6407±0.04240.6905±0.1942
40.8800±0.10330.6909±0.15640.8000±0.17060.8000±0.17060.8182±0.13360.3455±0.12330.7636±0.1666
51.0000±0.00000.9010±0.10560.8506±0.09210.8766±0.10890.8766±0.09020.6510±0.06110.8247±0.1057
60.9602±0.03940.9329±0.03150.9465±0.03870.9597±0.03560.9333±0.05150.8660±0.05250.9513±0.0411
70.8893±0.05880.8393±0.11750.7662±0.09940.8555±0.13090.8539±0.11940.1347±0.08140.8929±0.0911
80.9600±0.05160.9455±0.06560.8800±0.07480.9500±0.06710.7300±0.06400.6800±0.13270.9100±0.0943
90.9300±0.11600.8314±0.08430.8405±0.12830.8488±0.12360.8570±0.13140.6058±0.11800.5653±0.1443
100.9650±0.04090.8520±0.06750.6364±0.10810.8164±0.09400.7362±0.07310.7885±0.09100.6043±0.0787
110.8231±0.08920.4965±0.13670.4126±0.19200.3986±0.13850.3636±0.18290.4685±0.16550.4056±0.1576
121.0000±0.00000.9913±0.02740.9177±0.07890.9221±0.05840.7922±0.06510.9740±0.05510.6840±0.0819
), ArticleFig(id=1327012446352552086, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=EN, label=Table 5, caption=Classification Performance of R-FCCL and Six Popular Classification Methods, figureFileSmall=null, figureFileBig=null, tableContent=
序号R-FCCL(本文)SVMForestPAWiSARDSPAARCPCTBaggingILMPFTC
注:“±”前的数值为准确率,“±”后的数值为标准差.
11.0000±0.00000.6679±0.08720.9306±0.09850.8056±0.11100.8472±0.15720.9028±0.09490.7619±0.0286
20.8106±0.07990.4628±0.01750.7658±0.07830.6667±0.11660.6216±0.12060.6486±0.14890.6250±0.0257
30.9357±0.08330.6710±0.08660.7581±0.16140.7258±0.10180.7097±0.22080.7258±0.15880.5294±0.0429
40.8800±0.10330.7636±0.18720.7000±0.19440.8000±0.11350.6000±0.21080.6400±0.22710.8571±0.0361
51.0000±0.00000.6510±0.06110.9167±0.07380.7639±0.06320.9306±0.10100.9444±0.07380.8095±0.0192
60.9602±0.03940.9372±0.03780.9163±0.04610.7882±0.03190.8966±0.05450.8424±0.07720.9483±0.0090
70.8893±0.05880.8653±0.08140.7123±0.10120.7808±0.11880.6301±0.19550.6575±0.20920.8889±0.0180
80.9600±0.05160.8636±0.08810.8800±0.10330.9800±0.04220.6400±0.12650.7900±0.13700.9000±0.0225
90.9300±0.11600.8843±0.07230.8922±0.07170.8333±0.10510.8333±0.06790.7255±0.15720.8621±0.0200
100.9650±0.04090.9156±0.05170.7251±0.11510.8363±0.08320.5906±0.05810.5789±0.07960.4898±0.0227
110.8231±0.08920.0769±0.00000.7769±0.14710.6923±0.12460.6923±0.09590.7000±0.12270.2000±0.0163
121.0000±0.00000.9913±0.02740.9571±0.02700.9667±0.03920.8048±0.10400.8095±0.07781.0000±0.0000
), ArticleFig(id=1327012446428049559, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, language=CN, label=表5, caption=

R-FCCL与6种流行分类方法的分类性能对比分析

, figureFileSmall=null, figureFileBig=null, tableContent=
序号R-FCCL(本文)SVMForestPAWiSARDSPAARCPCTBaggingILMPFTC
注:“±”前的数值为准确率,“±”后的数值为标准差.
11.0000±0.00000.6679±0.08720.9306±0.09850.8056±0.11100.8472±0.15720.9028±0.09490.7619±0.0286
20.8106±0.07990.4628±0.01750.7658±0.07830.6667±0.11660.6216±0.12060.6486±0.14890.6250±0.0257
30.9357±0.08330.6710±0.08660.7581±0.16140.7258±0.10180.7097±0.22080.7258±0.15880.5294±0.0429
40.8800±0.10330.7636±0.18720.7000±0.19440.8000±0.11350.6000±0.21080.6400±0.22710.8571±0.0361
51.0000±0.00000.6510±0.06110.9167±0.07380.7639±0.06320.9306±0.10100.9444±0.07380.8095±0.0192
60.9602±0.03940.9372±0.03780.9163±0.04610.7882±0.03190.8966±0.05450.8424±0.07720.9483±0.0090
70.8893±0.05880.8653±0.08140.7123±0.10120.7808±0.11880.6301±0.19550.6575±0.20920.8889±0.0180
80.9600±0.05160.8636±0.08810.8800±0.10330.9800±0.04220.6400±0.12650.7900±0.13700.9000±0.0225
90.9300±0.11600.8843±0.07230.8922±0.07170.8333±0.10510.8333±0.06790.7255±0.15720.8621±0.0200
100.9650±0.04090.9156±0.05170.7251±0.11510.8363±0.08320.5906±0.05810.5789±0.07960.4898±0.0227
110.8231±0.08920.0769±0.00000.7769±0.14710.6923±0.12460.6923±0.09590.7000±0.12270.2000±0.0163
121.0000±0.00000.9913±0.02740.9571±0.02700.9667±0.03920.8048±0.10400.8095±0.07781.0000±0.0000
)], 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/issn1000-1239.202330428, detailUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202330428, pdfUrlCn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/PDF/10.7544/issn1000-1239.202330428, pdfUrlEn=https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/PDF/10.7544/issn1000-1239.202330428, aliStartDate=0, aliEndDate=0, collectionFlag=false, citedCount=null, citedUrl=null, previewStatus=0, delFlag=0, hasFullText=1, orderTime=1738339200000, fullTextJson=null, articleText=null, reference=null)
收藏切换
R-FCCL:一种面向高维数据的稳健模糊概念认知学习方法
郭豆豆 , 徐伟华 *
计算机研究与发展 | 人工智能 2025,62(2): 383-396
收起
收藏切换
计算机研究与发展 |人工智能 2025 , 62 (2) : 383 -396
R-FCCL:一种面向高维数据的稳健模糊概念认知学习方法
全屏
[Author(id=1327012443919855733, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=doudou876517690@126.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1327012443991158903, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012443919855733, language=EN, stringName=Doudou Guo, firstName=Doudou, middleName=null, lastName=Guo, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=College of Artificial Intelligence, Southwest University, Chongqing 400715, bio={"content":"

born in 1995. PhD candidate, Member of CCF. His main research interests include concept-cognitive learning, granular computing, and machine learning

"}, bioImg=null, bioContent=

born in 1995. PhD candidate, Member of CCF. His main research interests include concept-cognitive learning, granular computing, and machine learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012444070850680, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012443919855733, language=CN, stringName=郭豆豆, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=西南大学人工智能学院 重庆 400715, bio={"img":"kxPEdRGh/RpZzMHtPQh4Cw==","content":"

1995年生. 博士研究生. CCF会员. 主要研究方向为概念认知学习、粒计算、机器学习

"}, bioImg=kxPEdRGh/RpZzMHtPQh4Cw==, bioContent=

1995年生. 博士研究生. CCF会员. 主要研究方向为概念认知学习、粒计算、机器学习

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012443827581041, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, xref=null, ext=[AuthorCompanyExt(id=1327012443835969650, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Artificial Intelligence, Southwest University, Chongqing 400715), AuthorCompanyExt(id=1327012443848552563, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西南大学人工智能学院 重庆 400715)])]), Author(id=1327012444133765242, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=chxuwh@gmail.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1327012444217651324, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012444133765242, language=EN, stringName=Weihua Xu, firstName=Weihua, middleName=null, lastName=Xu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=College of Artificial Intelligence, Southwest University, Chongqing 400715, bio={"content":"

born in 1979. PhD, professor, PhD supervisor. His main research interests include granular computing, cognitive computing, data mining, and machine learning

"}, bioImg=null, bioContent=

born in 1979. PhD, professor, PhD supervisor. His main research interests include granular computing, cognitive computing, data mining, and machine learning

, aboutCorrespAuthor=null), CN=AuthorExt(id=1327012444288954493, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, authorId=1327012444133765242, language=CN, stringName=徐伟华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=西南大学人工智能学院 重庆 400715, bio={"img":"SdG9RUMTN6GtS4OA/Lu6xA==","content":"

1979年生. 博士,教授,博士生导师. 主要研究方向为粒计算、认知计算、数据挖掘、机器学习

"}, bioImg=SdG9RUMTN6GtS4OA/Lu6xA==, bioContent=

1979年生. 博士,教授,博士生导师. 主要研究方向为粒计算、认知计算、数据挖掘、机器学习

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1327012443827581041, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, xref=null, ext=[AuthorCompanyExt(id=1327012443835969650, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=College of Artificial Intelligence, Southwest University, Chongqing 400715), AuthorCompanyExt(id=1327012443848552563, tenantId=1146029695717560320, journalId=1324759914146529478, articleId=1327012441898201188, companyId=1327012443827581041, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=西南大学人工智能学院 重庆 400715)])])]
郭豆豆 , 徐伟华*
作者信息
  • 西南大学人工智能学院 重庆 400715
通讯作者:
徐伟华()
作者简介:

1995年生. 博士研究生. CCF会员. 主要研究方向为概念认知学习、粒计算、机器学习

1979年生. 博士,教授,博士生导师. 主要研究方向为粒计算、认知计算、数据挖掘、机器学习

R-FCCL: An Approach of Fuzzy-Based Concept-Cognitive Learning with Robustness for High-Dimensional Data
Doudou Guo , Weihua Xu*
Affiliations
  • College of Artificial Intelligence, Southwest University, Chongqing 400715
  • born in 1995. PhD candidate, Member of CCF. His main research interests include concept-cognitive learning, granular computing, and machine learning

    born in 1979. PhD, professor, PhD supervisor. His main research interests include granular computing, cognitive computing, data mining, and machine learning

出版时间: 2025-02-01 doi: 10.7544/issn1000-1239.202330428
文章导航
收藏切换

随着全球信息化的高速发展,高维数据挖掘与知识发现成为了人工智能领域亟待破解的科学问题之一. 然而,由于高维数据中样本的稀疏性与特征的冗余性,传统统计学模型和机器学习方法的泛化性和可解释性遇到极大的挑战. 为此,针对高维数据与知识弱演化能力之间不平衡的科学问题,利用三支概念求解复杂问题的认知机理,提出了一种新的概念建模方法,即稳健模糊概念认知学习(fuzzy-based concept-cognitive learning with robustness,R-FCCL). 首先,借助概念的最大相似性原则,建立了基于R-FCCL的高维数据分类系统,并从概念的角度出发,研究了高维数据的知识结构和认知学习机理. 进一步,利用模糊三支概念的正、负算子从2个不同的角度刻画了模糊环境概念认知学习过程,进而基于概念融合的模糊三支概念完成概念辨识和数据分类. 通过在12个真实数据集与12种分类方法的实验分析,验证所提方法具有较好的鲁棒性和有效性.

形式概念分析  /  概念认知学习  /  粒计算  /  高维数据  /  三支决策

With the rapid development of global informationization, data mining and knowledge discovery of high-dimensional data have been a hotspot in the field of artificial intelligence and data science. However, the sparse sample and redundant feature issues of high-dimensional data make it challenging to ensure the generalization and interpretability of traditional statistical models and machine learning methods. Hence, we present fuzzy-based concept-cognitive learning with robustness for the imbalance problem between high-dimensional data and weak knowledge evolution ability. The main idea is to explore the knowledge structure and cognitive learning mechanism of high-dimensional data from the concept perspective. We propose a high-dimensional data classification method based on the concept-cognitive learning mechanism in the fuzzy formal context. Furthermore, the cognitive learning process of fuzzy concepts is described from two different perspectives by the positive and negative cognitive learning operators of fuzzy three-way concepts. Finally, the fusion of fuzzy three-way concepts completes the task of concept identification and data classification. Extensive experiments performed on 12 real data sets compared with 12 state-of-the-art classification methods also verify the robustness and effectiveness of the proposed method. The considered framework can provide a convenient novel tool for high-dimensional data knowledge discovery research and fuzzy-based concept-cognitive learning.

formal concept analysis  /  concept-cognitive learning  /  granular computing  /  high-dimensional data  /  three-way decision
郭豆豆, 徐伟华. R-FCCL:一种面向高维数据的稳健模糊概念认知学习方法. 计算机研究与发展, 2025 , 62 (2) : 383 -396 . DOI: 10.7544/issn1000-1239.202330428
Doudou Guo, Weihua Xu. R-FCCL: An Approach of Fuzzy-Based Concept-Cognitive Learning with Robustness for High-Dimensional Data[J]. Journal of Computer Research and Development, 2025 , 62 (2) : 383 -396 . DOI: 10.7544/issn1000-1239.202330428
近年来,随着全球信息技术产业持续化推进,大数据逐渐上升为国家战略性资源. 利用大数据技术进行知识发现与智慧决策已经在机器学习、风险评估、图像处理、决策科学等诸多研究领域取得突飞猛进的发展[1]. 然而,由于高维数据存在稀疏低维表征特性,传统统计学模型和机器学习方法的泛化性和可解释性遇到了极大的挑战[2]. 从机器学习和数据挖掘的角度来看,面向高维数据的分类方法缺乏一定的泛化能力和可解释性. 主要原因在于:1)高维特征空间中,特征维数远远高于样本数,导致数据分布呈现非线性,致使难以获取具备泛化性能的分界面;2) “维数灾难”导致以经验数据为基础的训练方法失效;3)特征模型过度依赖后续分类器或评估指标. 因此,开展有效的高维数据的知识建模方法研究变得尤为重要.
在过去几十年间,不同分类场景下的分类器相继被提出,但其分类性能主要依赖于需要解决的分类场景,即数据集的范围和特定性质[3]. 正如“没有免费的午餐”定理(no free lunch theorem,NFT)[4]所言,不同分类器在处理相对应分类场景中各有所长,难分伯仲. 通俗来讲,分类器的设计往往需要依靠具有部分特性的分类场景来评估,例如:设计模糊分类器时往往考虑不平衡、高维等问题. 综合来看,优秀的分类器应具有3方面特性:1)特定分类场景下具备优秀的分类性能;2)合理的可解释性;3)良好的鲁棒性. 鉴于此,针对高维数据分类问题,本文结合三支概念求解复杂问题的认知机理,提出一种新的分类方法——模糊概念认知学习.
概念是人类在认识世界的过程中认识事物的本质特征,并进一步被高度抽象为获取的知识[5]. 借助概念,人们构建起一种精神与现实之间的映射,在辨识事物的同时,探索事物的发展规律[6],即认知. 概念通常由3方面组成,包括外延、内涵以及概念名,其中概念的外延是指概念描述所指代的个体范围,内涵是指概念的具体特征描述,而概念名即为其映射下内涵和外延的描述表示. 外延和内涵是唯一确定的,由外延和内涵相互指代的概念是独特的,因此可用于描述不同类型的事物[7]. 针对事物描述的语义解释,概念可以具体表示成形式概念[8]、粒概念[9]、近似概念[10]、面向对象概念[11]、面向属性概念[12]、双向概念[13]等,而且已经应用到不同的领域场景,例如知识发现[14]、机器学习[15]、数据挖掘[16-17]、云计算[18]等. 在概念的学习和认知方面,概念认知学习[9,19] (concept-cognitive learning,CCL)通过特定的认知方法完成概念内涵和外延的学习,即事物本质特征的识别. 注意到,三支概念[20-21]是形式概念分析领域中又一具有代表性的概念描述与分析理论,其特色在于同时从正、负这2种截然相反的角度去研究概念本身所蕴含的知识. 将三支概念理论用以描述模糊环境下的概念认知学习过程,可以有效缓解模糊环境中的不确定性,这也是本文采用三支概念进行认知学习的主要动机之一.
概念认知学习是以概念为知识的基本载体,研究事物认知学习过程的科学[22]. 从认知机制的视角来看,作为形式概念分析理论范式与粒计算理论的拓展,概念认知学习可同诸多认知科学分支相结合,从而产生新的认知计算方法. 比如同复杂网络结合产生了复杂网络下的概念认知学习[23],同模糊集方法相结合引发了模糊概念认知学习方法[24],同粒计算相结合演化了多水平概念认知学习[25]乃至多粒度概念认知学习[26]. 从学习机制的视角来看,作为概念学习和知识发现的核心工具,概念认知学习具备良好的事物辨识能力. 例如基于记忆的概念认知学习[27]和模糊粒概念认知学习[28]. 将概念作为基本出发点,借助概念之间的相似性公理能够很好地解决个案分类问题. 且主要优势体现在:1)可解释性. 由概念角度学习到的知识本身是因果知识[29-30],因此利用概念进行知识建模具有较强可解释性. 2)鲁棒性. 由内涵和外延相互指代的概念本身是独特的,能够精确地描述事物,结合概念之间相似性公理构建的分类器能取得较好的分类结果,且具有较强的鲁棒性. 目前,利用CCL开展机器学习的分类研究逐渐兴起,但对高维数据的概念建模与模糊分类问题尚未探究. 鉴于此,本文提出一种稳健的模糊概念认知学习方法,用以处理高维数据. 较之现有文献,有3个贡献:
1)提出一种面向高维数据的稳健模糊概念认知学习方法(fuzzy-based concept-cognitive learning method with robustness,R-FCCL);
2)在概念归类的一般性公理假设下,引入距离函数来构建相似类集合,并借助三支概念的正负认知算子描述模糊环境下对象与属性之间的关系,进而通过实验仿真验证R-FCCL学习方式的有效性;
3)结合模糊辨识关系和概念学习机制,设计了具有良好鲁棒性和可解释性的高维数据分类器.
本文提出的R-FCCL方法能在保证鲁棒性和可解释性的前提下具有一定的普适性,是一类针对高维数据的模糊分类器.
本节简要回顾模糊形式背景和概念认知学习的部分定义和性质,关于形式背景、模糊形式背景、概念格、三支概念分析等的详细性质及定义,可参阅相关书籍[31-33].
经典的形式背景是一类拥有特殊关系的数据库,通过对象集与属性集之间的等价二元关系,诱导出由对象集和属性集对形成的概念. 而模糊形式背景侧重将模糊集中隶属度的观点引入对象集和属性集之间的关系描述中,解决经典形式概念分析仅适用于离散型数据这一局限性. 需要强调的是,本文所讨论的概念认知学习均在模糊形式背景下建立.
给定一个论域集合$ U = \{ {x_1},{x_2}, … ,{x_n}\} $,论域$ U $上的模糊集$\tilde X$为
其中${\mu _{\tilde X}}:U \to [0,1]$,$ {\mu _{\tilde X}}(x) $表示对象$x$关于模糊集$X$的隶属度. 相应地,$\mu _{\tilde X}^{c} (x) = 1 - {\mu _{\tilde X}}(x)$表示其非隶属度,并用$ {\mathcal{T}^U} $表示定义在$U$上的所有模糊集的集合.
设五元组$(U,A,\tilde I,D,J)$为模糊形式背景,其中$U = \{ {x_1},{x_2}, … ,{x_n}\} $和$A = \{ {a_1},{a_2}, … ,{a_m}\} $分别表示对象集和条件属性集,$ \tilde I = \{ \langle (x,a),{\mu _I}(x,a)\rangle |(x,a) \in U \times A\} $表示对象集$U$与条件属性集$A$之间的模糊二元关系;$U/D = \{ {U_1},{U_2}, … ,{U_l}\} $,其中$l$表示决策类别总数;$J:U \times D \to \{ 0,1\} $为$U$与决策类集合$D$之间的二元关系. 此外,任意$(x,a) \in \tilde I$存在隶属度${\mu _{\tilde I}}(x,a) \in [0,1]$. 为了方便,记${\mu _{\tilde I}}(x,a)$为$\tilde I(x,a)$.
因此,给定模糊形式背景$(U,A,\tilde I,D,J)$,且$X \subseteq U$,$B \subseteq A$,$ \tilde B \in {\mathcal{T}^A} $,可定义一组算子:
其中${\mathcal{T}^A}$表示所有定义在$A$上的模糊集的集合. 此外,以上一组算子满足如下性质[30]:
1)${X_1} \subseteq {X_2} \Rightarrow X_2^* \subseteq X_1^*,{\tilde B_1} \subseteq {\tilde B_2} \Rightarrow \tilde B_2^* \subseteq \tilde B_1^*$;
2)$X \subseteq {X^{**}},\tilde B \subseteq {\tilde B^{**}}$;
3)$X = {X^{***}},\tilde B = {\tilde B^{***}}$;
4)$X \subseteq {\tilde B^*} \Leftrightarrow \tilde B \subseteq {X^*}$;
5)${({X_1} \cup {X_2})^*} = X_1^* \cap X_2^*,({\tilde B_1} \cup {\tilde B_2})^* = \tilde B_1^* \cap \tilde B_2^*$;
6)${({X_1} \cap {X_2})^*} \supseteq X_1^* \cup X_2^*,({\tilde B_1} \cap {\tilde B_2})^* \supseteq \tilde B_1^* \cup \tilde B_2^*$.
进而,若二元组$(X,\tilde B)$满足${X^*} = \tilde B$且${\tilde B^*} = X$,则称$(X,\tilde B)$为模糊概念,其中$X$表示模糊概念的外延,$\tilde B$表示模糊概念的内涵. 显然,$({X^{**}},{X^*})$和$({\tilde B^*},{\tilde B^{**}})$均为模糊概念. 用$\tilde L(U,A,\tilde I)$表示模糊形式背景中的所有模糊概念集合. 对任意$ ({X_1},{\tilde B_1}),({X_2},{\tilde B_2}) $ $ \in \tilde L(U,A,\tilde I) $,定义二者之间的序关系为$ ({X_1},{\tilde B_1}) $ $ \leqslant ({X_2},{\tilde B_2}) \Leftrightarrow {X_1} $$ \subseteq {X_2} \Leftrightarrow {\text{ }}{\tilde B_2} \subseteq {\tilde B_1} $. 此时,称$({X_1},{\tilde B_1})$为$({X_2},{\tilde B_2})$的子概念,或$({X_2},{\tilde B_2})$为$({X_1},{\tilde B_1})$的父概念.
基于经典形式背景的CCL强调一种非此即彼描述的概念学习,忽略了具有连续性数据形式背景的概念表示与学习. 基于模糊形式背景的CCL侧重于对象与模糊属性之间的隶属关系. 因此,模糊概念认知学习适合应用于连续型数据场景的概念学习.
设$(U,A,\tilde I,D,J)$为模糊形式背景,${2^U}$和${\mathcal{T}^A}$分别表示对象集合和属性集合的幂域,2个集值映射$ \tilde {\mathcal{F}} $和$\mathcal{H}$分别表示为$\tilde {\mathcal{F}}:{2^U} \to {\mathcal{T}^A}$和$\mathcal{H}:{\mathcal{T}^A} \to {2^U}$,${L_1} = P(U)$和${\tilde L_2} = P(A)$ 分别表示完备格和模糊完备格. 设$\tilde L$表示格,${0_{\tilde L}}$和${1_{\tilde L}}$分别表示相应的零元和单位元.
定义1. 设$(U,A,\tilde I,D,J)$为模糊形式背景,${L_1}$和${\tilde L_2}$是一对完备格,对任意${X_1},{X_2} \in {L_1}$,称$\tilde {\mathcal{F}}:{L_1} \to {\tilde L_2}$是外延-内涵认知学习算子,若其满足2个性质:
1)$\tilde {\mathcal{F}}({0_{{L_1}}}) = {1_{{{\tilde L}_2}}},\tilde {\mathcal{F}}({1_{{L_1}}}) = {0_{{{\tilde L}_2}}}$;
2)$\tilde {\mathcal{F}}({X_1} \vee {X_2}) = \tilde {\mathcal{F}}({X_1}) \wedge \tilde {\mathcal{F}}({X_2})$.
类似地,对任意${\tilde B_1},{\tilde B_2} \in {\text{ }}{\tilde L_2}$,称$ \mathcal{H}:{\text{ }}{\tilde L_2} \to {L_1} $是内涵-外延认知学习算子,如果其满足2个性质:
1)$\mathcal{H}({0_{{L_2}}}) = {1_{{{\tilde L}_1}}},\mathcal{H}({1_{{L_2}}}) = {0_{{{\tilde L}_1}}}$;
2)$\mathcal{H}({\tilde B_1} \vee {\tilde B_2}) = \mathcal{H}({\tilde B_1}) \wedge \mathcal{H}({\tilde B_2})$.
上述概念认知算子$\tilde {\mathcal{F}}$和$\mathcal{H}$具备形式概念分析中算子具有的性质,相对于形式概念分析,概念认知学习算子对概念学习的约束更为宽松[9,20].
定义 2. 设$(U,A,\tilde I,D,J)$为模糊形式背景,对任意$X \subseteq U$,正外延-内涵认知学习算子$\tilde {\mathcal{F}}:{2^U} \to {\mathcal{T}^A}$与正内涵-外延认知学习算子$\mathcal{H}:{\mathcal{T}^A} \to {2^U}$定义如下:
类似地,负外延-内涵认知学习算子${\tilde {\mathcal{F}}^ - }:{2^U} \to {\mathcal{T}^A}$与负内涵-外延认知学习算子${\mathcal{H}^ - }:{\mathcal{T}^A} \to {2^U}$定义为:
其中${\tilde I^ - } = \{ \langle (x,a),1 - {\mu _{\tilde I(x,a)}}\rangle |(x,a) \in U \times A\} $为$\tilde I$的补集,表示其负隶属关系.
定义3. 设$(U,A,\tilde I,D,J)$为模糊形式背景,对于任意$X \subseteq U$以及$ {\tilde B_1},{\tilde B_2} \in {\mathcal{T}^A} $,模糊三支概念认知学习算子${\tilde {\mathcal{F}}^\nabla }:{2^U} \to {\mathcal{T}^A} \times {\mathcal{T}^A}$和${\mathcal{H}^\nabla }:{\mathcal{T}^A} \times {\mathcal{T}^A} \to {2^U}$分别定义为:
当${\tilde {\mathcal{F}}^\nabla }(X) = ({\tilde B_1},{\tilde B_2})$且${\mathcal{H}^\nabla }({\tilde B_1},{\tilde B_2}) = X$成立时,称$(X,({\tilde B_1},{\tilde B_2}))$为模糊三支概念. 此外,当$(X,({\tilde B_1},{\tilde B_2}))$ $ \leqslant (X',({\tilde B'_1},{\tilde B'_2}))$成立时,称$(X',({\tilde B'_1},{\tilde B'_2}))$是$(X,({\tilde B_1},{\tilde B_2}))$的父概念,即$X \subseteq X'$且$({\tilde B'_1},{\tilde B'_2}) \leqslant ({\tilde B_1},{\tilde B_2})$成立. 若$(X',({\tilde B'_1},{\tilde B'_2}))$仅有自身1个父概念,则称其为最大父概念. 而本质上,定义2表示序对$(x,a)$和$\tilde I$在$a$上的隶属度,以及$(x,a)$和$\tilde I$在$a$上的非隶属度. 因此,正、负认知算子具备相同的性质.
由上可知,模糊三支概念同时从正、负概念认知算子2个角度刻画出对象集和模糊属性集之间的关系. 相较于其他概念表示形式,三支概念往往对事物本质描述更加详细. 更多关于模糊三支概念和模糊三支算子的定义和性质可参阅文献[34].
高维数据分类是以历史经验数据进行机理化建模的关键步骤,也是当前大数据和人工智能领域的热点课题. 本节以设计高维数据的分类器为研究主题,并以构建具备良好分类性能、可解释性、鲁棒性的分类方法为目标,提出一类可用于高维数据的模糊分类方法. 众所周知,关于概念归类的一般性公理,为概念相似性假设和概念识别提供了理论基础. 其核心思想来源于文献[6],具体描述如下:
公理1. 样本可分性公理. 任意对象总存在一个唯一的类与其最相似.
公理2. 类别可分性公理. 任意类至少存在一个对象与其最相似.
公理3. 归类可分性公理. 任意类的认知表示与外延表示的归类能力等价.
不难发现,上述一般性公理的主要出发点是相似性准则,即认为一个对象可以被归类为相似性程度最大的类别中. 而概念恰恰是从内涵和外延2个方面来刻画,即认知表示和外延表示. 因此,借助归类的一般性公理,可以完成概念的认知学习与辨别.
由于高维数据的稀疏性和冗余性,其分类模型通常难以具有较强的泛化能力. 为此,R-FCCL 考虑从部分关键特征出发学习概念,进而利用概念之间的相似性完成个案分类. 为提取关键特征进行概念学习,本节根据相似类中的对象相似性与不同相似类中的对象分离性构建模糊辨识关系并设计相应的属性重要性度量指标.
设$(U,A,\tilde I,D,J)$为模糊形式背景,且${\tilde {\mathcal{F}}^d} :{2^U} \to {2^D}$,${\mathcal{H}^d} : {2^D} \to {2^U}$为2个集值映射. 对任意$x \in U$,将所有决策概念$({\mathcal{H}^d}{\tilde {\mathcal{F}}^d}(x),{\text{ }}{\tilde {\mathcal{F}}^d}(x))$记为${\mathcal{G}^d}$ $ = \{ ({\mathcal{H}^d}{\tilde {\mathcal{F}}^d}(x),{\tilde {\mathcal{F}}^d}(x))|x \in U\} $,称${\mathcal{G}^d}$为决策概念空间,其设定具有类内相似性原则,即相似对象间无差异.
定义4. 给定模糊形式背景$(U,A,\tilde I,D,J)$,对于任意${x_i},{x_j} \in U,a \in A$,对象${x_i}$和${x_j}$关于特征$a$的模糊辨识关系定义为:
其中阈值$\varepsilon \in [0,1)$为无差异系数,$re{c_a} :U \times U \to [0,1]$. ${\mathcal{G}^d}({x_i}) = {\mathcal{G}^d}({x_j})$表示同一概念空间的2个对象,即具有相同决策类标签. 当二者间距离小于等于$\varepsilon $时,可认为2个对象间是无差异的;反之,认为2个对象间存在差异性. 显然,$re{c_a}({x_i},{x_i}) = 0$和$re{c_a}({x_i},{x_j}) = re{c_a}({x_j},{x_i})$成立. 特别地,若$\varepsilon = 1$,则所有对象间的距离均小于等于$\varepsilon $,缺乏模糊辨识能力,可设置模糊辨识度为0.
由上可知,定义4 给定了3种情况下的模糊辨识关系:1)处于同一概念空间的对象之间无差异,记模糊辨识度为0;2)相似对象之间无差异,记模糊辨识度为0;3)对象间距离大于设定的无差异系数,记模糊辨识度大于0,且距离越远,相应的模糊辨识度越高.
任意${x_i},{x_j} \in U$关于特征$a \in A$和特征集合$A' \subseteq A$的模糊辨识度为:
由上述分析不难发现,特征$a$的模糊辨识度$re{c_a}$反映了$a$对所有对象的总体区分能力,特征集合$A'$的模糊辨识程度$re{c_{A'}}({x_i},{x_j})$反映出$A'$对${x_i}$和${x_j}$的区分能力.
定义5. 给定模糊形式背景$(U,A,\tilde I,D,J)$,称$R \subseteq A$为模糊辨识关系的关键特征集合,特征$a \in R$是关键特征,若$R$与$a$满足2个性质:
1)$\forall {x_i},\,\,{x_j} \in U,\,\,\exists a \in R,\,\,{\text{s}}{\text{.t}}{\text{.}}\, re{c_a}({x_i},{x_j}) \gt 0 $, if $re{c_A}({x_i}, {x_j})\ne 0$.
2)$\forall a \in R,\exists {x_i},{x_j} \in U,{\text{s}}{\text{.t}}{\text{. }} re{c_{R - a}}({x_i},{x_j}) \gt 0$, if $re{c_A}({x_i}, {x_j}) \ne 0$.
具体而言,性质1)保证了$U$中的任意2个对象,总存在至少1个关键特征用于区分这2个对象;性质2)保证了关键特征集合中任意特征在对象辨识过程中发挥相应作用,而减小关键特征集合的冗余性.
进一步地,可以根据关键特征集合进行概念认知学习.
定义6. 设$(U,R,\tilde I,D,J)$为模糊形式背景,$R \subseteq A$是关键特征集合,$U/D = \{ {U_1},{U_2}, … ,{U_l}\} $是对象集合$U$在决策属性$D$下的决策类划分. 对于任意$x \in {U_i}\,(i = 1,2, … ,l)$,${a_j} \in R\,(j = 1,2, … ,|R|)$,对象$x$在关键特征集$R$下的相似类定义为距离对象$x$最近的$k$个对象的集合,记作${N_k}(x)$,距离$d$计算定义为:
其中$\tilde I(x,{a_j})$和${\tilde I^ - }(x,{a_j})$分别表示隶属度和非隶属度.
由于相似类中对象之间是联系紧密的,故可通过对象$x$的相似类集合${N_k}(x)$进行概念学习,且有如下定理.
定理1. 设五元组$(U,R,\tilde I,D,J)$为模糊形式背景,其中,$R \subseteq A$是关键特征集合,$U/D = \{ {U_1},{U_2}, … ,{U_l}\} $表示决策类划分. 给定任意相似类集合${N_k}(x)$,则$(\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x)),(\tilde {\mathcal{F}}({N_k}(x)),{\tilde {\mathcal{F}}^ - }({N_k}(x))))$为对象导出的模糊三支概念.
证明. 要证明定理1成立,仅需证明 $(\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x)),(\tilde {\mathcal{F}}({N_k}(x)),{\tilde {\mathcal{F}}^ - }({N_k}(x))))$满足定义3 中模糊三支概念认知算子的2条性质即可,即${\mathcal{H}^\nabla }(\tilde {\mathcal{F}}({N_k}(x)), {\tilde {\mathcal{F}}^ - }({N_k}(x))) = \mathcal{H}\tilde {\mathcal{F}} ({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x))$及${\tilde {\mathcal{F}}^\nabla } (\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x))) = (\tilde {\mathcal{F}}({N_k}(x)),{\tilde {\mathcal{F}}^ - }({N_k}(x))) $.
1)根据定义3可直接证得性质1)成立,即$ {\mathcal{H}^\nabla }(\tilde {\mathcal{F}}({N_k}(x)), {\tilde {\mathcal{F}}^ - }({N_k}(x))) = \mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x)) $ .
2)要证性质2)成立,需证3步:
①由$\tilde {\mathcal{F}}({N_k}(x))(a) = \mathop \wedge \limits_{x \in {N_k}(x)} \tilde I(x,a),\;{\tilde {\mathcal{F}}^ - }({N_k}(x))(a) = \mathop \wedge \limits_{x \in {N_k}(x)} {\tilde I^ - }(x,a)$可得$ \mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) = \{ x \in U|\tilde I(x,a) \geqslant \mathop \wedge \limits_{x \in {N_k}(x)} \tilde I(x,a), a \in R\} $成立,且
故$ \mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x)) = \{ x \in U|\mathop \wedge \limits_{x \in {N_k}(x)} \tilde I(x,a) \leqslant\tilde I(x, a) \leqslant \mathop \vee \limits_{x \in {N_k}(x)} \tilde I(x,a),a \in R\} ,$
并且$\tilde {\mathcal{F}}(\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x))(a)) = \mathop \wedge \limits_{x \in {N_k}(x)} \tilde I(x,a)= \tilde {\mathcal{F}} ({N_k}(x))(a) $成立.
②由于
则$ {\tilde {\mathcal{F}}^ - }(\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x))(a)){\text{ }} = \mathop \wedge \limits_{x \in {N_k}(x)} {\tilde I^c}(x,a) = {\tilde {\mathcal{F}}^ - }({N_k}(x))(a) $成立.
因此,${\tilde {\mathcal{F}}^\nabla }(\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x))) = (\tilde {\mathcal{F}}({N_k}(x)),{\text{ }}$ ${\tilde {\mathcal{F}}^ - }({N_k}(x)))$成立. 证毕.
定理2. 设$(U,R,\tilde I,D,J)$为模糊形式背景,且$(\mathcal{H},\tilde {\mathcal{F}})$和$({\mathcal{H}^ - },{\tilde {\mathcal{F}}^ - })$是2对认知算子. 给定相似类${N_k}(x) \subseteq U$,则有2个结论成立.
1)$\tilde {\mathcal{F}}({N_k}(x)) = \mathop \cap \limits_{y \in {N_k}(x)} \tilde {\mathcal{F}}(y) $,
  $ {{\tilde {\mathcal{F}}}^ - }({N_k}(x)) = \mathop \cap \limits_{y \in {N_k}(x)} {{\tilde {\mathcal{F}}}^ - }(y) $.
2)$\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) = \mathop \cap \limits_{a \in \tilde {\mathcal{F}}({N_k}(x))} \mathcal{H}(a) $,
  $ {\mathcal{H}^ - }{{\tilde {\mathcal{F}}}^ - }({N_k}(x)) = \mathop \cap \limits_{a \in {{\tilde {\mathcal{F}}}^ - }({N_k}(x))} {\mathcal{H}^ - }(a) $.
证明. 由定义2、定义3及定理1直接可证. 证毕.
由定理2可知,R-FCCL是依托$k$值获取相似类集合,并以此为基础进行概念学习. 而$k$值的设定关系到模糊概念的学习,会影响后续的分类进程. 因此,参数$k$的学习在R-FCCL 系统中十分重要. 在此基础上,算法1 给出了面向高维数据的概念学习全过程.
算法1. 概念学习.
输入:模糊形式背景$(U,A,\tilde I,D,J)$,$\varepsilon $,$k$;
输出:模糊三支概念.
① 令$R = \varnothing $,$\tilde {\mathcal{G}} = \varnothing $,${\tilde {\mathcal{G}}_i} = \varnothing $;
② for $a \in A$
③  for ${x_i},{x_j} \in U$
④  根据定义4计算对象间辨识度$re{c_a}({x_i},{x_j})$;
⑤  计算$a$与$R$的模糊辨识度$re{c_a}$和$re{c_R}({x_i},{x_j})$;
⑥  end for
⑦ end for
⑧ while $\displaystyle\sum_{a \in A} {re{c_a}} \ne 0$
⑨  ${a^*} = \mathop {\arg {\text{max}}}\limits_{a \in A - R} re{c_a}$ ;
⑩  if $re{c_{{a^*}({x_i},{x_j})}} \gt 0$
⑪   $R = R \cup \{ {a^*}\} $;
⑫    for $({x_i},{x_j}) \in U \times U$
⑬     if $re{c_{a*}}({x_i},{x_j}) \gt 0$
⑭      $re{c_a}({x_i},{x_j}) = 0,\forall a \in A$;
⑮     end if
⑯    end for
⑰  end if
⑱ end while
⑲ for ${U_i} \in U/D$
⑳  for$x \in {U_i}$
㉑   根据定义6获取${N_k}(x)$;
㉒   基于定理1获取模糊三支概念 $ (\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{{\tilde {\mathcal{F}}}^ - }({N_k}(x)),(\tilde {\mathcal{F}}({N_k}(x)) $, ${{\tilde {\mathcal{F}}}^ - }({N_k}(x)))) $;
㉓   $ {{\tilde {\mathcal{G}}}_i} \leftarrow {{\tilde {\mathcal{G}}}_i} \cup (\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{{\tilde {\mathcal{F}}}^ - }({N_k}(x)) $, $(\tilde {\mathcal{F}}({N_k}(x))),({{\tilde {\mathcal{F}}}^ - }({N_k}(x)))) $;
㉔  end for
㉕ end for
例 1. 表1为一个模糊形式背景,包括9个对象和5个属性. 记对象集$U = \{ {x_1},{x_2}, … ,{x_9}\} $,条件属性集$A = \{ {a_1},{a_2}, … ,{a_4}\} $,决策属性$D = \{ d\} $且$U/D = \{ {U_1},{U_2}\} $.
给定无差异系数$\varepsilon = 0.3$,定义4所有对象在4个属性上的模糊辨识度计算为:
根据辨识度最大原则,属性${a_2}$首先被选为关键特征. 对于任意对象${x_i},{x_j} \in U$,若其满足$re{c_{{a_2}}}({x_i},{x_j}) \gt 0$,则记$re{c_a}({x_i},{x_j}) = 0,\forall a \in A$. 此时,可计算模糊辨识度为:
重复上述过程. 易得关键特征集$R = \{ {a_2},{a_4}\} . $
根据定义6计算对象${x_i} \in U$在其决策类下的相似类${N_k}(x)(K = 3)$为:
根据定理1学习模糊三支概念为:
从不同相似类学习到的模糊三支概念所包含的信息量均不相同,即概念辨识能力不同. 为了进一步提高模糊概念在后续概念识别中的识别能力,文献[14]提出了一种基于模糊概念聚类的方法来生成模糊概念本体,文献[34]研究了一种基于渐进式三支概念的方法来实现模糊概念本体的生成. 本文在这些方法的基础上,提出一种模糊三支概念融合方法来进一步提升概念识别能力.
给定$(U,R,\tilde I,D,J)$模糊形式背景,$\tilde {\mathcal{F}}:{2^U} \to {\mathcal{T}^R}$,$\mathcal{H}:{\mathcal{T}^R} \to {2^U}$和${\tilde {\mathcal{F}}^ - }:{2^U} \to {\mathcal{T}^R}$,${\mathcal{H}^ - }:{\mathcal{T}^R} \to {2^U}$为2对正、负认知算子. 由定理1可知,对于任意$x \in U$,$a \in R$,$(\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x)),(\tilde {\mathcal{F}}({N_k}(x)),{\tilde {\mathcal{F}}^ - }({N_k}(x))))$为对象导出模糊三支概念,记模糊三支概念子空间为${\tilde {\mathcal{G}}_i} = \{ (\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{\tilde {\mathcal{F}}^ - }({N_k}(x)),(\tilde {\mathcal{F}}({N_k}(x)),{\tilde {\mathcal{F}}^ - }({N_k}(x))))$$|x \in {U_i}\} . $为方便表示,下文中将模糊三支概念简记为$({S_\lambda },({\tilde B_\lambda }({S_\lambda }),\tilde B_\lambda ^ - ({S_\lambda })))$,其中$\lambda = 1,2, … ,j$.
定义7. $({S_1},({\tilde B_1}({S_1}),\tilde B_1^ - ({S_1}))),({S_2},({\tilde B_2}({S_2}),\tilde B_2^ - ({S_2}))),… ,$ $ ({S_{ j}},({\tilde B_j}({S_{ j}}),\tilde B_j^ - ({S_{ j}})))$为一簇模糊三支概念,若其满足$\forall {S_i} \subseteq {S_{ j}}(i \leqslant j)$,则称该模糊三支概念簇可融合;记$({S_u},({\tilde B_u}({S_u}),\tilde B_u^ - ({S_u})))$为融合后的模糊三支概念,其融合过程定义为:
由上可知,不同概念在概念融合过程中所占权重不同,权重随着外延增大而增大,即概念外延泛化能力越强在概念融合过程中起的影响越大. 从认知角度来看,定义7完全可以刻画出一种符合人类认知的渐进式概念认知过程[6]. 同时,基于定义7给出的模糊三支概念融合方法,可以对任意模糊三支概念子空间${\tilde {\mathcal{G}}_i}$进行概念融合,记概念融合后的子空间为$\tilde {\mathcal{G}}_i^F = \{ ({S_u},({\tilde B_u}({S_u}),\tilde B_u^ - ({S_u})))|{S_u} \subseteq {U_i}\} $. 显然,任意模糊三支概念$({S_u},({\tilde B_u}({S_u}),\tilde B_u^ - ({S_u}))) \in {\text{ }}\tilde {\mathcal{G}}_i^F$有2个性质成立.
1)对任意融合后的模糊三支概念$ (S_u,(\tilde{B}_u(S_u), $ $\tilde B_u^ - ({S_u}))) \in \tilde {\mathcal{G}}_i^F$,仅有自身1个父概念;
2)对任意模糊三支概念$({S_{{\lambda _j}}},({\tilde B_{{\lambda _j}}}({S_{{\lambda _j}}}),\tilde B_{{\lambda _j}}^ - ({S_{{\lambda _j}}}))) \in $${\tilde {\mathcal{G}}_i}$,至少存在1个融合后的模糊三支概念的外延${S_u}$包含${S_{{\lambda _j}}}$,即${S_{{\lambda _j}}} \subseteq {S_u}$.
定理3. 设$(U,R,\tilde I,D,J)$为模糊形式背景,${\tilde {\mathcal{G}}_i}$和$\tilde {\mathcal{G}}_i^F$分别表示概念融合前后的概念子空间,则有如下结论成立:
证明. 为证明定理3成立,仅需证明3步:
1)由定义7可知,对于任意${S_i} \subseteq {\tilde {\mathcal{G}}_i}$,若存在${S_u} \subseteq {\tilde {\mathcal{G}}_i}$使得${S_i} \subseteq {S_u}$,则有$ \left| {\tilde {\mathcal{G}}_i^F} \right| = 1 $成立.
2)由定义7可知,对于任意${S_i} \subseteq {\tilde {\mathcal{G}}_i}$,若不存在${S_u} \subseteq {\tilde {\mathcal{G}}_i}$使得${S_i} \subseteq {S_u}$,则有$ \left| {\tilde {\mathcal{G}}_i^F} \right| = \left| {{{\tilde {\mathcal{G}}}_i}} \right| $成立.
3)对于任意${S_i},{S_u} \subseteq {\tilde {\mathcal{G}}_i}$,若存在${S_i} \subseteq {S_u}$,则有$1 \lt \left| {\tilde {\mathcal{G}}_i^F} \right| \lt \left| {{{\tilde {\mathcal{G}}}_i}} \right|$成立. 证毕.
由上可知,定义7从认知角度刻画出一个全新的模糊三支概念. 尽管此概念是以伪概念的形式呈现,但此概念是后续概念识别过程中的一个中间变量. 本文不进行伪概念与概念二者之间的区分,伪模糊三支概念亦称之为模糊三支概念. 基于上述讨论,算法2给出了概念融合的过程.
算法2. 概念融合.
输入:模糊形式背景$(U,A,\tilde I,D,J)$,$\varepsilon $,$k$;
输出:模糊三支概念空间${\tilde {\mathcal{G}}^F}$.
① 基于算法1学习模糊三支概念;
② for ${U_i} \in U/D$
③  初始化${\tilde {\mathcal{G}}_i} = \varnothing $;
④  for$x \in {U_i}$
⑤   根据定义6获取${N_k}(x)$;
⑥   根据定理1获取模糊三支概念    $ (\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{{\tilde {\mathcal{F}}}^ - }({N_k}(x)) $,    $ (\tilde {\mathcal{F}}({N_k}(x)),{{\tilde {\mathcal{F}}}^ - }({N_k}(x)))) $;
⑦    $ {{\tilde {\mathcal{G}}}_i} \leftarrow {{\tilde {\mathcal{G}}}_i} \cup (\mathcal{H}\tilde {\mathcal{F}}({N_k}(x)) \cap {\mathcal{H}^ - }{{\tilde {\mathcal{F}}}^ - }({N_k}(x)) $, $(\tilde {\mathcal{F}}({N_k}(x))),({{\tilde {\mathcal{F}}}^ - }({N_k}(x)))) $;
⑧  end for
⑨ end for
⑩ for ${\tilde {\mathcal{G}}_i} \in \tilde {\mathcal{G}}$
⑪  $\tilde {\mathcal{G}}_i^F = \varnothing $;
⑫  for $({S_{{\lambda _u}}},({\tilde B_{{\lambda _u}}}({S_{{\lambda _u}}}),\tilde B_{{\lambda _u}}^ - ({S_{{\lambda _u}}}))) \in {\tilde {\mathcal{G}}_i}$
⑬   根据定义7获取模糊三支概念融合 $({S_u},({\tilde B_u}({S_u}),{\text{ }}\tilde B_u^ - ({S_u})))$;
⑭   $\tilde {\mathcal{G}}_i^F \leftarrow \tilde {\mathcal{G}}_i^F \cup ({S_u},({\tilde B_u}({S_u}),\tilde B_u^ - ({S_u})))$;
⑮   ${\tilde {\mathcal{G}}^F} \leftarrow {\tilde {\mathcal{G}}^F} \cup {\text{ }}\tilde {\mathcal{G}}_i^F$;
⑯  end for
⑰ end for
例 2.(接例1)模糊三支概念子空间${\tilde {\mathcal{G}}_1}$中存在的2个可融合模糊三支概念簇分别如下:
和
根据定义7,可得
由此可得融合后模糊三支概念子空间${\tilde {\mathcal{G}}_1}^F$如下:
同理,可以计算模糊三支概念子空间${\tilde {\mathcal{G}}_2}^F$如下:
基于2.2节所述的概念融合与概念空间构建方法,本节进一步探讨一种基于概念辨别的分类机制.
定义8. 设$U' = \{ {x'_1},{x'_2}, … ,{x'_l}\} $为待识别对象集合,$ {\tilde {\mathcal{G}}^F} = \{ \tilde {\mathcal{G}}_1^F,\tilde {\mathcal{G}}_2^F, … ,\tilde {\mathcal{G}}_l^F\} $为融合后的模糊三支概念空间,对任意模糊三支概念$({S_u},({\tilde B_u}({S_u}),\tilde B_u^ - ({S_u}))) \in \tilde {\mathcal{G}}_i^F$,待识别对象$x' \in U'$与此概念之间的辨识程度可定义为
其中$\tilde B$和${\tilde B^ - }$分别表示待识别对象在关键特征下的隶属度和非隶属度.
由样本可分性公理可知,任意对象总存在唯一类与其最相似,且二者间相似性程度越低,其辨识度越高. 因此,对于任意模糊三支概念子空间$\tilde {\mathcal{G}}_i^F$,易得该子空间内与待识别对象$x' \in U'$辨识程度最小的模糊三支概念,其辨识度记为$Rec(x',\tilde {\mathcal{G}}_i^F({S_u}))$. 进而对于整体概念空间${\tilde {\mathcal{G}}^F}$,易得与待识别对象辨识度最低的模糊三支概念子空间,即待识别对象$x' \in U'$的归类,记为${k^*}$,其计算方式为:
事实上,概念辨识过程就是对象归类过程,关键在于如何构建模糊三支概念空间和获取最小辨识度的模糊三支概念所在空间.
例3. 给定一个待识别对象$ (\{{x}_{0}\}, (\{\langle {a}_{2},0.28\rangle ,\langle {a}_{4}, 0.48\rangle \},\{\langle {a}_{2},0.72\rangle ,\langle {a}_{4},0.52\rangle \})), $可计算其与不同概念间的辨识程度为:
由此可确定待识别对象${x_0}$与模糊三支概念子空间${\tilde {\mathcal{G}}_1}^F$最相似.
故判定待识别对象${x_0}$的标签为$d = 1$.
根据上述分析,算法3给出了概念辨识全过程.
算法3. 概念辨识.
输入:模糊三支概念空间${\tilde {\mathcal{G}}^F}$,待识别个案$x'$;
输出:概念标签${k^*}$.
① 令$R = \varnothing $,$\tilde {\mathcal{G}} = \varnothing $;
②  计算概念辨识;
③ for $({S_u},({\tilde B_u}({S_u}),\tilde B_u^ - ({S_u}))) \in {\tilde {\mathcal{G}}^F}$
④  根据定义8计算$x'$与任意融合概念的辨识 度$Rec(x',{\tilde {\mathcal{G}}^F}({S_u}))$;
⑤ end for
⑥ ${k^*} = \mathop {\arg {\text{ }}min}\limits_{i \in \{ 1,2, … ,l\} } Rec(x',{\text{ }}\tilde {\mathcal{G}}_i^F({S_u}))$.
综上,R-FCCL 方法的理论已全部建立,其过程描述如算法1~3所示. 基于第2节讨论可知,本文方法主要由3阶段组成,即概念学习、概念融合与概念辨识,其对应的算法1~3的时间复杂度分别为$O(|U{|^2}|A|)$,$O(|\tilde {\mathcal{G}}{|^2})$,$O(|{\tilde {\mathcal{G}}^F}{|^2})$,其中$|\tilde {\mathcal{G}}|$和$|{\tilde {\mathcal{G}}^F}{|^2}$分别表示相应融合前后模糊三支概念空间中的概念基数. 由2.1节和2.2节分析可知,模糊三支概念空间中的概念数小于等于样本数,故易知算法时间复杂度为$O(|U{|^2}|A|)$. 然而,当高维空间中样本数远远小于特征维数时,算法时间复杂度为$O(|A|)$.
通过在公开数据集上实验,本节将所提出的R-FCCL方法与多种主流分类算法进行对比,分别从分类性能、参数敏感度、鲁棒性3个方面分析所提方法的有效性. 具体实验平台如下:操作系统为Microsoft Win10;处理器为Intel® CoreTM i7-10750H CPU 2.60 GHz;存储器为32 GB;编程语言为MATLAB 2020a.
实验中共选取了12个真实数据集[35],详情如表2所示,采用10折交叉验证方法来评估R-FCCL 的分类性能. 由于源数据集不能直接用于模糊背景的实验验证,故实验将选取的数据集均按文献[36]方法对条件属性进行模糊化处理:
其中$v(x,a)$表示$x$在$a$上的取值,${V_{\max }}(x,a)$和${V_{\min }}(x,a)$分别表示所有对象在属性$a$上的最大值和最小值.
同时需要注意到,$\varepsilon $与$k$均是关乎分类系统的重要参数,且$\varepsilon $的值域为$[0,1]$,在具体实验中设定$\varepsilon $的步长为0.1,即$\varepsilon \in \{ 0,0.1, … ,1\} $;将$k$的值域范围设置为1~10,并设定步长为1,即$k \in \{ 1,2, … ,10\} $. 进一步地,为充分说明R-FCCL方法的有效性,本文选取了12种主流分类方法与之对比,详情如表3所示.
根据所选对比算法的特点,可将这些算法大致分为2类进行单独对比与分析. 第1类是6种基于模糊最近邻分类算法,包括KNN(K=3),FENN,IFKNN,PFKNN,FRNN-FRS,CFNN等算法. 第2类是6种其他先进分类算法,包括SVM,ForestPA,WiSARD,SPAARC,PCTBagging,ILMPFTC等算法. 鉴于R-FCCL方法的理论基础为概念归类的一般性公理,因此第1类方法主要是选取基于模糊集理论的最近邻分类方法,而第2类方法则采用当前比较先进的分类算法.
R-FCCL与第1类对比算法在12个数据集上的详细结果(包括平均分类正确率和标准差)如表4所示. 观察表4可以发现,R-FCCL在高维数据集上表现出优秀的分类性能. 与第1类对比算法相比,除第7个数据集Lung_Discrete以外,R-FCCL 在剩下的11个数据集上均取得了最好的分类准确率,其中在第1,5,12个数据集,采用R-FCCL 的分类准确率可以达到100%. 此外,WiSARD方法在Orlraws10P数据集上可以取得最好的分类准确率,ILMPFTC在WarpPIE10P上的分类准确率也可以达到100%. 而与第2类对比算法在所选数据集的详细结果如表5 所示,不难发现,与这6种先进的分类方法相比,R-FCCL仍然能在除Orlraws10P以外的11个数据集上表现出优秀的分类准确率.
因此,从表4和表5的整体情况来看,R-FCCL在分类准确率方面优于其他所选的12种流行的对比方法,具备更好的泛化性能. 此外,图1直观反映了不同分类方法在分类准确率和标准差方面的对比情况.
由第2节可知,R-FCCL方法的2个核心参数是$\varepsilon $与$k$,因此本节在12个所选数据集上对R-FCCL进行了参数敏感性检验,各组参数组合均采用10折交叉验证并记录其分类准确率结果如图2 所示. 在图2中,通过比较不同参数组合下R-FCCL在12个所选数据集下的分类准确率,可以清楚观察其变化趋势. 此外,根据参数的敏感性程度可以充分说明不同的参数对于系统的重要性不同,也说明不同数据集下选择合适的参数十分必要. 这恰好验证了引言部分提到的分类器受场景的影响较大.
为进一步综合验证R-FCCL与不同分类方法在不同数据集上的分类表现,本节将对其进行鲁棒性分析,给出鲁棒性的定义[47]如下:
对于任意方法${\mathcal{M}_i}(i \in 1,2, … ,m)$,在数据集$D{S_{ j}}(j = 1,2, … ,s)$上的鲁棒性为其方法${\mathcal{M}_i}$在数据集$D{S_{ j}}$上的准确率与所有方法中的最小准确率的比值,即
其中$ac{c_{{\mathcal{M}_i}}}(D{S_{ j}})$为方法${\mathcal{M}_i}$在数据集$D{S_{ j}}$上的准确率,${\text{min }}ac{c_{{\mathcal{M}_i}}}(D{S_{ j}})$为所有方法在数据集$D{S_{ j}}$上的最小准确率.
由上可知,方法${\mathcal{M}_i}$的总体鲁棒性为该方法在所有数据集的鲁棒性之和,即${r_{{\mathcal{M}_i}}} = \displaystyle\sum_{j = 1}^s {{r_{{\mathcal{M}_i}}}} (D{S_{ j}})$. 此外,需要说明的是,鲁棒性值越大说明该方法性能越好. 图3 展示了R-FCCL分别与2类对比方法在12 个数据集的鲁棒性对比. 通过图3可以清楚观察到R-FCCL的总体鲁棒性在所有对比算法中排名第1,体现出R-FCCL在高维数据分类问题上的有效性.
概念认知学习是以概念为知识的基本载体,研究事物认知学习过程的科学. 其基本思路是通过特定的认知方法来学习概念并以此研究事物认知学习过程. 其研究可追溯到张文修教授和徐伟华教授[9]的早期文章. 受此文启发,陆续有不少学者加入相关方向研究. 近年来,随着概念认知学习理论在模型推广、方法设计和实际应用等方面的不断深入,越来越多的研究者开始接受和接触这一新兴理论. 特别是,概念认知学习与机器学习二者富有成效地结合也极大地开阔这一新兴课题的研究视野.
针对高维数据分类问题,本文提出了模糊概念认知学习系统,旨在借助概念来刻画数据的知识结构和认知机理,并以此完成分类. 具体而言,在模糊形式背景下,构建了一种带有鲁棒性的模糊概念认知学习系统,用于高维数据的认知学习建模和分类问题解决. 其主要步骤包括:1)提出基于模糊辨识关系的关键特征概念学习方法;2)构建基于概念融合的模糊三支概念空间;3)借助概念归类的一般性公理完成概念识别. 此外,实验分析部分通过与12种主流分类方法对比,验证了所提方法的有效性.
本文主要探讨了如何应用模糊概念认知学习方法到高维数据分类问题,给高维数据分类和概念认知学习研究提供新的思路和方法. 研究结果体现了概念认知学习是一类有效的数据知识发现方法. 然而,R-FCCL主要关注于如何进行高维数据的概念建模和以及如果增强高维数据分类性能,并没有讨论如何降低概念学习的复杂性以及大规模高维数据分类等问题. 此外,在本文研究基础上,仍有许多问题有待深入研究,例如大规模高维数据的概念认知学习方法、动态数据实时更新的认知学习概念方法、并行计算意义下的概念认知学习机制、模糊概念认知学习有效性的衡量指标和评判标准等. 这些课题的研究将进一步提高概念认知学习的有效性和实用性,从而为解决知识驱动的大数据智能分析问题及可解释性人工智能研究奠定基础.
作者贡献声明:郭豆豆落实完善理论方法、设计实验方案、分析对比实验,并撰写修改论文;徐伟华提出理论思想、算法思路、指导意见,并审核论文.

参考文献 引证文献
排序方式:
1
于洪,何德牛,王国胤,等. 大数据智能决策[J]. 自动化学报,2020,46(5):878−896
Yu Hong, He Deniu, Wang Guoyin, et al. Big data for intelligent decision making[J]. Acta Automatica Sinica, 2020, 46(5): 878−896 (in Chinese)
2
于晓,刘慧,吴彦,等. 基于本质自表示的多视角子空间聚类[J]. 中国科学:信息科学,2021,51(10):1625−1639
Yu Xiao, Liu Hui, Wu Yan, et al. Intrinsic self-representation for multi-view subspace clustering[J]. SCIENTIA SINICA Informations, 2021, 51(10): 1625−1639 (in Chinese)
3
Cózar J, Fernández A, Herrera F, et al. A metahierarchical rule decision system to design robust fuzzy classifiers based on data complexity[J]. IEEE Transactions on Fuzzy Systems, 2019, 27(4): 701−715
4
Gómez D, Rojas A. An empirical overview of the no free lunch theorem and its effect on real-world machine learning classification[J]. Neural Computing, 2016, 28(1): 216−228
5
Lake B M, Salakhutdinov R, Tenenbaum J B, et al. Human-level concept learning through probabilistic program induction[J]. Science, 2015, 350(6266): 1332−1338
6
于剑. 机器学习:从公理到算法[M]. 北京:清华大学出版社,2017
Yu Jian. Machine Learning from Axioms to Algorithms[M]. Beijing: Tsinghua University Press, 2017 (in Chinese)
7
李金海,米允龙,刘文奇. 概念的渐进式认知理论与方法[J]. 计算机学报,2019,42(10):2233−2250
Li Jinhai, Mi Yunlong, Liu Wenqi. Incremental cognitive of concept: Theories and methods[J]. Chinese Journal of Computers, 2019, 42(10): 2233−2250 (in Chinese)
8
李金海,李玉斐,米允龙,等. 多粒度形式概念分析的介粒度标记方法[J]. 计算机研究与发展,2020,57(2):447−458
Li Jinhai, Li Yufei, Mi Yunlong, et al. Meso-granularity labeled method for multi-granularity formal concept analysis[J]. Journal of Computer Research and Development, 2020, 57(2): 447−458 (in Chinese)
9
张文修,徐伟华. 基于粒计算的认知模型[J]. 工程数学学报,2007,24(6):957−971
Zhang Wenxiu, Xu Weihua. Cognitive model based on granular computing[J]. Chinese Journal of Engineering Mathematics, 2007, 24(6): 957−971 (in Chinese)
10
Li Jinhai, Mei Changlin, Xu Weihua, et al. Concept learning via granular computing: A cognitive viewpoint[J]. Information Sciences, 2015, 298: 447−467
11
王霞,江山,李俊余,等. 三元概念的一种构造方法[J]. 计算机研究与发展,2019,56(4):844−853
Wang Xia, Jiang Shan, Li Junyu, et al. A construction method of triadic concepts[J]. Journal of Computer Research and Development, 2019, 56(4): 844−853 (in Chinese)
12
Yao Yiyu. Interval sets and three-way concept analysis in incomplete contexts[J]. International Journal of Machine Learning and Cybernetics, 2017, 8: 3−20
13
Xu Weihua, Guo Doudou, Mi Jusheng, et al. Two-way concept-cognitive learning via concept movement viewpoint[J]. IEEE Transactions on Neural Networks and Learning Systems. 2023, 34(10): 6798−6812
14
Mi Yunlong, Shi Yong, Li Jinhai, et al. Fuzzy-based concept learning method: Exploiting data with fuzzy conceptual clustering[J]. IEEE Transactions on Cybernetics, 2021, 51(2): 809−821
15
Shi Yong, Mi Yunlong, Li Jinhai, et al. Concurrent concept-cognitive learning model for classification[J]. Information Sciences, 2019, 496: 65−81
16
Zhang Xiaoyan, Guo Doudou, Xu Weihua. Two-way concept-cognitive learning with multi-source fuzzy context[J]. Cognitive Computation, 2023, 15(5): 1526−1548
17
Yan Enliang, Yu Cunguo, Lu Liming, et al. Incremental concept cognitive learning based on three-way partial order structure[J]. Knowledge-Based Systems, 2021, 220: 68−98
18
Hao Fei, Pang Guangyao, Pei Zheng, et al. Virtual machines scheduling in mobile edge computing: A formal concept analysis approach[J]. IEEE Transactions on Sustainable Computing, 2019, 5(3): 319−328
19
魏玲,曹丽,祁建军,等. 形式概念分析中的概念约简与概念特征[J]. 中国科学:信息科学,2020,50(12):1817−1833
Wei Ling, Cao Li, Qi Jianjun, et al. Concept reduction and concept characteristics in formal concept analysis[J]. SCIENTIA SINICA Informations, 2020, 50(12): 1817−1833 (in Chinese)
20
徐伟华,杨蕾,张晓燕. 模糊三支形式概念分析与概念认知学习[J]. 西北大学学报:自然科学版,2020,50(4):516−528
Xu Weihua, Yang Lei, Zhang Xiaoyan. Fuzzy three-way formal concept analysis and concept-cognitive learning[J]. Journal of Northwest University: Natural Science Edition, 2020, 50(4): 516−528 (in Chinese)
21
Qian Ting, Wei Ling, Qi Jianjun. Constructing three-way concept lattices based on apposition and subposition of formal contexts[J]. Knowledge-Based Systems, 2017, 116(15): 39−48
22
Guo Doudou, Xu Weihua. Fuzzy-based concept-cognitive learning: An investigation of novel approach to tumor diagnosis analysis[J]. Information Sciences, 2023, 639: 118998
23
马娜,范敏,李金海. 复杂网络下的概念认知学习[J]. 南京大学学报:自然科学,2019,55(4):609−623
Ma Na, Fan Min, Li Jinhai. Concept-cognitive learning under complex network[J]. Journal of Nanjing University: Natural Science, 2019, 55(4): 609−623 (in Chinese)
24
Xu Weihua, Guo Doudou, Qian Yuhua, et al. Two-way concept-cognitive learning method: A fuzzy-based progressive learning[J]. IEEE Transactions on Fuzzy Systems, 2023, 31(6): 1885−1899
25
Fan Bingjiao, Tsang E C C, Xu Weihua, et al. Attribute-oriented cognitive concept learning strategy: A multi-level method[J]. International Journal of Machine Learning and Cybernetics, 2019, 10(9): 2421−2437
26
Li Jinhai, Huang Chenchen, Qi Jianjun, et al. Three-way cognitive concept learning via multi-granularity[J]. Information Sciences, 2017, 378: 244−263
27
Guo Doudou, Xu Weihua, Qian Yuhua, et al. Fuzzy-granular concept-cognitive learning via three-way decision: Performance evaluation on dynamic knowledge discovery[J/OL]. IEEE Transactions on Fuzzy Systems, 2023[2023-10-19]. https://ieeexplore.ieee.org/document/10288039
28
Guo Doudou, Xu Weihua, Qian Yuhua, et al. M-FCCL: Memory-based concept-cognitive learning for dynamic fuzzy data classification and knowledge fusion[J]. Information Fusion, 2023, 100: 101962
29
Zhang Tao, Rong Mei, Shan Haoran, et al. Causal asymmetry analysis in the view of concept-cognitive learning by incremental concept tree[J]. Cognitive Computation, 2021, 13(5): 1274−1286
30
Xu Weihua, Li Wentao. Granular computing approach to two-way learning based on formal concept analysis in fuzzy dataset[J]. IEEE Transactions on Cybernetics, 2016, 46(2): 66−379
31
Ganter B. Formal Concept Analysis: Mathematical Foundations[M]. Berlin: Springer, 1999
32
徐伟华,李金海,魏玲,等. 形式概念分析理论与应用[M]. 北京:科学出版社,2016
Xu Weihua, Li Jinhai, Wei Ling, et al. Formal Concept Analysis: Theory and Application[M]. Beijing: Science Press, 2016 (in Chinese)
33
祁建军,魏玲,姚一豫. 三支概念分析与决策[M]. 北京:科学出版社,2019
Qi Jianjun, Wei Ling, Yao Yiyu. Three-Way Concept Analysis and Decision[M]. Beijing: Science Press, 2019 (in Chinese)
34
Yuan Kehua, Xu Weihua, Li Wentao, et al. An incremental learning mechanism for object classification based on progressive fuzzy three-way concept[J]. Information Sciences, 2022, 584: 127−147
35
Li Jundong, Cheng Kewei, Wang Suhang, et al. Feature selection: A data perspective[J]. ACM Computing Surveys, 2018, 50(6): 94−139
36
Elhadad A, Ghareeb A, Abbas S. A blind and high-capacity data hiding of DICOM medical images based on fuzzification concepts[J]. Alexandria Engineering Journal, 2021, 66(2): 2471−2482
37
Kuncheva L I. An intuitionistic fuzzy k-nearest neighbors rule[J]. Notes on Intuitionistic Fuzzy Sets, 1995, 1: 56−60
38
Yang M S, Cheng C H. On the edited fuzzy k-nearest neighbor rule[J]. IEEE Transactions on Systems, Man, and Cybernetics: Cybernetics, 1998, 28(8): 461−466
39
Arif M, Akram M U, Afsar F A, et al. Pruned fuzzy k-nearest neighbor classifier for beat classification[J]. Journal of Biomedical Science and Engineering, 2010, 3(4): 380−389
40
Jensen R, Cornelis C. Fuzzy-rough nearest neighbour classification[C] // Proc of the 7th Rough Sets and Current Trends in Computing Conf. Berlin: Springer, 2011: 56−72
41
Zhai Junhai, Li Na, Zhai Mengyao. The condensed fuzzy k-nearest neighbor rule based on sample fuzzy entropy[C] // Proc of the 24th Int Conf on Machine Learning and Cybernetics. Berlin: Springer, 2011: 282−286
42
周志华. 机器学习[M]. 北京:清华大学出版社,2016
Zhou Zhihua. Machine Learning[M]. Beijing: Tsinghua University Press, 2016 (in Chinese)
43
Adnan M N, Islam M Z. Forest PA: Constructing a decision forest by penalizing attributes used in previous trees[J]. Expert Systems with Applications, 2017, 89(15): 389−403
44
De G M, Maurizio G. An experimental evaluation of weightless neural networks for multi-class classification[J]. Applied Soft Computing, 2018, 72: 338−354
45
Yates D, Islam M Z, Gao J. SPAARC: A Fast Decision Tree Algorithm[M]. Berlin: Springer, 2019
46
Ii A, Jmp B, Jm B, et al. PCTBagging: From inner ensembles to ensembles. A trade-off between discriminating capacity and interpretability [J]. Information Sciences, 2022, 583: 219−238
47
Vlachos M, Domeniconi C, Gunopulos D, et al. Non-linear dimensionality reduction techniques for classification and visualization[C]//Proc of the 8th ACM Int Conf on Knowledge Discovery and Data Mining. New York: ACM, 2002: 645−651
2025年第62卷第2期
110
48
0
引用本文
BibTeX
文章信息
doi: 10.7544/issn1000-1239.202330428
  • 接收时间:2023-06-05
  • 首发时间:2026-11-10
  • 出版时间:2025-02-01
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2023-06-05
  • 修回日期:2024-02-06
基金
作者信息
    西南大学人工智能学院 重庆 400715

通讯作者:

徐伟华()
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/jsjyjyfz/CN/10.7544/issn1000-1239.202330428
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