Article(id=1153429494946844900, tenantId=1146029695717560320, journalId=1149652044408987649, issueId=1153429493357203682, articleNumber=null, orderNo=null, doi=10.19812/j.cnki.jfsq11-5956/ts.20241008004, pmid=null, cstr=null, oa=null, hot=null, price=null, onlineType=0, articleFormat=0, articleType=null, articleTypeStr=null, receivedDate=1728316800000, receivedDateStr=2024-10-08, revisedDate=null, revisedDateStr=null, acceptedDate=null, acceptedDateStr=null, onlineDate=1752928621278, onlineDateStr=2025-07-19, pubDate=1741968000000, pubDateStr=2025-03-15, doiRegisterDate=null, doiRegisterDateStr=null, onlineIssueDate=1752928621278, onlineIssueDateStr=2025-07-19, onlineJustAcceptDate=null, onlineJustAcceptDateStr=null, onlineFirstDate=null, onlineFirstDateStr=null, sourceXml=null, magXml=null, createTime=1752928621278, creator=13701087609, updateTime=1752928621278, updator=13701087609, issue=Issue{id=1153429493357203682, tenantId=1146029695717560320, journalId=1149652044408987649, year='2025', volume='16', issue='5', pageStart='1', pageEnd='326', issueExtLink='null', onlineDate='null', pubDate='null', beforeIssueId=null, nextIssueId=null, price=null, status=1, issueComplete=1, articleOrder=1, issueType=-1, specialIssue=0, createTime=1752928620900, creator=13701087609, updateTime=1758690311058, updator=13701087609, preIssue=null, nextIssue=null, ext={EN=IssueExt(id=1177595773500932351, tenantId=1146029695717560320, journalId=1149652044408987649, issueId=1153429493357203682, language=EN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=), CN=IssueExt(id=1177595773500932352, tenantId=1146029695717560320, journalId=1149652044408987649, issueId=1153429493357203682, language=CN, specialIssueTitle=, coverIllustrator=null, specialIssueEditor=, specialIssueAbout=)}, issueFiles=null}, startPage=187, endPage=196, ext={EN=ArticleExt(id=1153429495353692391, articleId=1153429494946844900, tenantId=1146029695717560320, journalId=1149652044408987649, language=EN, title=Stacked generalization model prediction of fenthion spot-check results in vegetables based on particle swarm optimization ensemble learning algorithm, columnId=1153429495274000613, journalTitle=Journal of Food Safety & Quality, columnName=Special Topic: Research and Detection of Pesticide and Veterinary Drug Residue, runingTitle=null, highlight=null, articleAbstract=

Objective To establish a vegetable safety risk prediction model based on the particle swarm optimization (PSO) algorithm and the stacked generalization (Stacking) model, and to predict the sampling results of fenthion in vegetables sold in Shanghai. Methods Based on the sampling data of fenthion in vegetables sold in Shanghai from 2021 to 2023, task type, sampling area, sampling link, sampling place, sampling month, testing institution, and vegetable variety were selected as feature variables. The target variable was whether the sampling results for fenthion in vegetables were qualified. The PSO-Stacking prediction model was constructed using ten-fold cross-validation to select effective machine learning models and resampling methods and optimized the model parameters using the PSO algorithm. Results Fenthion-positive samples were found in 55 out of 3889 vegetable samples, with an overall failure rate of 1.4%. Bean vegetables had the highest rate at 2.3%, followed by eggplant and fruiting vegetables at 0.2%. The base models were obtained through screening, including Random Forest (RF), categorical boosting (CatBoost), gradient boosting (GB), extreme gradient Boosting (XGBoost), and light gradient boosting machine (LGBM). The best resampling technique was adaptive synthetic sampling (ADASYN). The PSO-Stacking model achieved the highest precision (0.91), recall (0.83), F1 score (0.87), and area under the curve (AUC) value (0.91) on the test set. Conclusion The PSO-Stacking model effectively addresses imbalanced food safety sampling data, accurately predicts the unqualified fenthion samples in vegetables, and provides technical support for vegetable supervision, sampling and risk warning.

, correspAuthors=Shao-Jie PENG, authorNote=null, correspAuthorsNote=null, copyrightStatement=null, 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, authorCompany=null, fund=null, authors=null, authorsList=Zi-Wen ZHOU, Zhi-Yi FAN, Shao-Jie PENG), CN=ArticleExt(id=1153429527003911095, articleId=1153429494946844900, tenantId=1146029695717560320, journalId=1149652044408987649, language=CN, title=基于粒子群优化集成学习算法堆叠模型预测蔬菜中倍硫磷的抽检结果, columnId=1153429495479521513, journalTitle=食品安全质量检测学报, columnName=本期专题:农兽药残留研究与检测, runingTitle=null, highlight=null, articleAbstract=

目的 建立基于粒子群(particle swarm optimization, PSO)算法优化堆叠模型(stacked generalization, Stacking)的蔬菜安全风险预测模型, 对上海市市售蔬菜中倍硫磷的抽检结果进行预测。方法 基于2021—2023年上海市市售蔬菜中倍硫磷的抽检数据, 选取任务类型、抽样地区、抽样环节、抽样场所、抽样月份、检测机构、蔬菜品种作为特征变量, 以蔬菜中倍硫磷的抽检结果是否合格为目标变量; 采用十折交叉验证筛选优良机器学习模型、重采样方法, 经PSO算法优化模型参数后构建PSO-Stacking预测模型。结果 3889件蔬菜中检出倍硫磷阳性样品55件, 不合格率为1.4%。其中, 豆类蔬菜不合格率最高(2.3%), 其次为茄果类(0.2%)。筛选得到基模型, 包括随机森林(random forest, RF)、类别特征梯度提升树(categorical boosting, CatBoost)、梯度提升(gradient boosting, GB)、极端梯度提升(extreme gradient boosting, XGBoost)和轻量级梯度提升机(light gradient boosting machine, LGBM), 最佳重采样方法为自适应合成抽样(adaptive synthetic sampling, ADASYN)技术。PSO-Stacking模型在测试集上的精确率(0.91)、召回率(0.83)、F1值(0.87)和曲线下面积(area under the curve, AUC)值(0.91)均为最高。结论 PSO-Stacking模型在不均衡食品安全抽检数据中表现优异, 能准确预测蔬菜中倍硫磷不合格样本, 为蔬菜监督抽检及风险预警提供技术支撑。

, correspAuthors=彭少杰, authorNote=null, correspAuthorsNote=
* 彭少杰(1968—), 男, 硕士, 主任医师, 主要研究方向为食品安全风险监测与评估。E-mail:
, copyrightStatement=null, copyrightOwner=null, extLink=null, articleAbsUrl=null, sourceXml=dMRZVI8GRPAHckF8kzQsmw==, magXml=DMERpRKmfGacPsKNW5911A==, pdfUrl=null, pdf=Xq9ug4nULBvZJegNyrVt6g==, pdfFileSize=null, pdfExtLink=null, richHtmlUrl=null, mobilePdfUrl=null, reviewReport=null, pdfFirstPage=null, abstractGraph=null, abstractGraphContent=null, abstractVideo=null, citation=null, cebUrl=null, magXmlContent=ctcSXgU2u8p6kQig3zLGJg==, mapNumber=null, authorCompany=null, fund=null, authors=

周子文(1999—), 男, 硕士, 主要研究方向为食品质量与安全(风险监测与评估)。E-mail:

, authorsList=周子文, 范志仪, 彭少杰)}, authors=[Author(id=1177619618865754261, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=zzw1174600720@163.com, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1177619619117412506, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, authorId=1177619618865754261, language=EN, stringName=Zi-Wen ZHOU, firstName=Zi-Wen, middleName=null, lastName=ZHOU, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1177619619184521370, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, authorId=1177619618865754261, language=CN, stringName=周子文, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=上海市市场监督管理局信息应用研究中心, 上海 200030, bio={"content":"

周子文(1999—), 男, 硕士, 主要研究方向为食品质量与安全(风险监测与评估)。E-mail:

"}, bioImg=null, bioContent=

周子文(1999—), 男, 硕士, 主要研究方向为食品质量与安全(风险监测与评估)。E-mail:

, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1177619618744119440, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, xref=null, ext=[AuthorCompanyExt(id=1177619618752508048, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China), AuthorCompanyExt(id=1177619618760896657, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海市市场监督管理局信息应用研究中心, 上海 200030)])]), Author(id=1177619619318739104, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1177619619427791014, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, authorId=1177619619318739104, language=EN, stringName=Zhi-Yi FAN, firstName=Zhi-Yi, middleName=null, lastName=FAN, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1177619619536842923, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, authorId=1177619619318739104, language=CN, stringName=范志仪, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=上海市市场监督管理局信息应用研究中心, 上海 200030, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1177619618744119440, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, xref=null, ext=[AuthorCompanyExt(id=1177619618752508048, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China), AuthorCompanyExt(id=1177619618760896657, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海市市场监督管理局信息应用研究中心, 上海 200030)])]), Author(id=1177619619629117617, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, orderNo=2, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=pengshaojie2008@qq.com, emailSecond=null, emailThird=null, correspondingAuthor=1, authorType=1, ext={EN=AuthorExt(id=1177619619817861302, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, authorId=1177619619629117617, language=EN, stringName=Shao-Jie PENG, firstName=Shao-Jie, middleName=null, lastName=PENG, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1177619619926913211, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, authorId=1177619619629117617, language=CN, stringName=彭少杰, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=*, address=上海市市场监督管理局信息应用研究中心, 上海 200030, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1177619618744119440, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, xref=null, ext=[AuthorCompanyExt(id=1177619618752508048, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China), AuthorCompanyExt(id=1177619618760896657, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海市市场监督管理局信息应用研究中心, 上海 200030)])])], keywords=[Keyword(id=1177619620283429057, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, orderNo=1, keyword=vegetables), Keyword(id=1177619620493144260, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, orderNo=2, keyword=fenthion), Keyword(id=1177619620681887943, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, orderNo=3, keyword=particle swarm algorithm), Keyword(id=1177619620774162634, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, orderNo=4, keyword=stacked generalization model), Keyword(id=1177619620866437327, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, orderNo=5, keyword=machine learning), Keyword(id=1177619620967100626, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, orderNo=6, keyword=food safety), Keyword(id=1177619621076152532, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, orderNo=1, keyword=蔬菜), Keyword(id=1177619621269090519, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, orderNo=2, keyword=倍硫磷), Keyword(id=1177619621369753818, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, orderNo=3, keyword=粒子群算法), Keyword(id=1177619621428474075, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, orderNo=4, keyword=堆叠模型), Keyword(id=1177619621516554464, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, orderNo=5, keyword=机器学习), Keyword(id=1177619621587857632, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, orderNo=6, keyword=食品安全)], refs=[Reference(id=1177619624666476825, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2016, volume=30, issue=9, pageStart=1744, pageEnd=1753, url=null, language=null, rfNumber=[1], rfOrder=0, authorNames=王芳, 李兆周, 李道敏, journalName=核农学报, refType=null, unstructuredReference=王芳, 李兆周, 李道敏, 等. 倍硫磷表面印迹聚合物的制备及表征[J]. 核农学报, 2016, 30(9): 1744-1753., articleTitle=倍硫磷表面印迹聚合物的制备及表征, refAbstract=null), Reference(id=1177619624758751513, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2016, volume=30, issue=9, pageStart=1744, pageEnd=1753, url=null, language=null, rfNumber=[1], rfOrder=1, authorNames=WANG F, LI ZZ, LI DM, journalName=Journal of Nuclear Agricultural Sciences, refType=null, unstructuredReference=WANG F, LI ZZ, LI DM, et al. Preparation and characterization of fenthion surface molecularly imprinted polymers[J]. Journal of Nuclear Agricultural Sciences, 2016, 30(9): 1744-1753., articleTitle=Preparation and characterization of fenthion surface molecularly imprinted polymers, refAbstract=null), Reference(id=1177619624880386331, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=2, authorNames=王子浩, journalName=脂肪乳联合血液灌流对血浆中倍硫磷清除效果的评价, refType=null, unstructuredReference=王子浩. 脂肪乳联合血液灌流对血浆中倍硫磷清除效果的评价[D]. 长春: 吉林大学, 2024., articleTitle=null, refAbstract=null), Reference(id=1177619624972661020, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[2], rfOrder=3, authorNames=WANG ZH, journalName=The evaluation of the clearance effect of fenthion in plasma by combining fat emulsion with hemoperfusion, refType=null, unstructuredReference=WANG ZH. The evaluation of the clearance effect of fenthion in plasma by combining fat emulsion with hemoperfusion[D]. Changchun: Jilin University, 2024., articleTitle=null, refAbstract=null), Reference(id=1177619625064935709, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2017, volume=36, issue=9, pageStart=1099, pageEnd=1103, url=null, language=null, rfNumber=[3], rfOrder=4, authorNames=刘津, 甘兰萍, 孙通, journalName=分析测试学报, refType=null, unstructuredReference=刘津, 甘兰萍, 孙通, 等. 激光诱导击穿光谱结合CARS变量选择方法定量检测倍硫磷含量[J]. 分析测试学报, 2017, 36(9): 1099-1103., articleTitle=激光诱导击穿光谱结合CARS变量选择方法定量检测倍硫磷含量, refAbstract=null), Reference(id=1177619625119461663, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2017, volume=36, issue=9, pageStart=1099, pageEnd=1103, url=null, language=null, rfNumber=[3], rfOrder=5, authorNames=LIU J, GAN LP, SUN T, journalName=Journal of Instrumental Analysis, refType=null, unstructuredReference=LIU J, GAN LP, SUN T, et al. Quantitative determination of fenthion using laser induced breakdown spectroscopy with CARS variable selection method[J]. Journal of Instrumental Analysis, 2017, 36(9): 1099-1103., articleTitle=Quantitative determination of fenthion using laser induced breakdown spectroscopy with CARS variable selection method, refAbstract=null), Reference(id=1177619625228513569, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=23, issue=1, pageStart=70, pageEnd=74, url=null, language=null, rfNumber=[4], rfOrder=6, authorNames=步岩刚, 钱宗耀, journalName=现代农药, refType=null, unstructuredReference=步岩刚, 钱宗耀. 倍硫磷在新疆哈密瓜中的残留检测及其在土壤中的降解行为研究[J]. 现代农药, 2024, 23(1): 70-74., articleTitle=倍硫磷在新疆哈密瓜中的残留检测及其在土壤中的降解行为研究, refAbstract=null), Reference(id=1177619625383702818, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=23, issue=1, pageStart=70, pageEnd=74, url=null, language=null, rfNumber=[4], rfOrder=7, authorNames=BU YG, QIAN ZY, journalName=Modern Agrochemicals, refType=null, unstructuredReference=BU YG, QIAN ZY. Detection of fenthion residues in Hami melon and its degradation behavior in soils[J]. Modern Agrochemicals, 2024, 23(1): 70-74., articleTitle=Detection of fenthion residues in Hami melon and its degradation behavior in soils, refAbstract=null), Reference(id=1177619625446617379, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=8, authorNames=吴帅, journalName=面向不平衡数据的食品安全风险预警技术研究, refType=null, unstructuredReference=吴帅. 面向不平衡数据的食品安全风险预警技术研究[D]. 南宁: 广西大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619625610195237, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[5], rfOrder=9, authorNames=WU S, journalName=Research on food safety risk prediction technology based on unbalanced data, refType=null, unstructuredReference=WU S. Research on food safety risk prediction technology based on unbalanced data[D]. Nanning: Guangxi University, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619625736024359, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=43, issue=12, pageStart=2549, pageEnd=2561, url=null, language=null, rfNumber=[6], rfOrder=10, authorNames=WANG X, BOUZEMBRAK Y, OUDE-LANSINK AGJM, journalName=Risk Analysis, refType=null, unstructuredReference=WANG X, BOUZEMBRAK Y, OUDE-LANSINK AGJM, et al. Weighted Bayesian network for the classification of unbalanced food safety data: Case study of risk-based monitoring of heavy metals[J]. Risk Analysis, 2023, 43(12): 2549-2561., articleTitle=Weighted Bayesian network for the classification of unbalanced food safety data: Case study of risk-based monitoring of heavy metals, refAbstract=null), Reference(id=1177619625874436393, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=150, issue=null, pageStart=111103, pageEnd=null, url=null, language=null, rfNumber=[7], rfOrder=11, authorNames=LI T, ZUO E, CHEN C, journalName=Applied Soft Computing, refType=null, unstructuredReference=LI T, ZUO E, CHEN C, et al. Gaussian distribution resampling via Chebyshev distance for food computing[J]. Applied Soft Computing, 2024, 150: 111103., articleTitle=Gaussian distribution resampling via Chebyshev distance for food computing, refAbstract=null), Reference(id=1177619625945739563, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=50, issue=24, pageStart=353, pageEnd=361, url=null, language=null, rfNumber=[8], rfOrder=12, authorNames=丁浩晗, 田嘉伟, 谢祯奇, journalName=食品与发酵工业, refType=null, unstructuredReference=丁浩晗, 田嘉伟, 谢祯奇, 等. 机器学习和大数据在食品领域的应用[J]. 食品与发酵工业, 2024, 50(24): 353-361., articleTitle=机器学习和大数据在食品领域的应用, refAbstract=null), Reference(id=1177619626054791469, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=50, issue=24, pageStart=353, pageEnd=361, url=null, language=null, rfNumber=[8], rfOrder=13, authorNames=DING HH, TIAN JW, XIE ZQ, journalName=Food and Fermentation Industries, refType=null, unstructuredReference=DING HH, TIAN JW, XIE ZQ, et al. Applications of machine learning and big data in food industry[J]. Food and Fermentation Industries, 2024, 50(24): 353-361., articleTitle=Applications of machine learning and big data in food industry, refAbstract=null), Reference(id=1177619626159649071, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2020, volume=14, issue=2, pageStart=241, pageEnd=258, url=null, language=null, rfNumber=[9], rfOrder=14, authorNames=DONG X, YU Z, CAO W, journalName=Frontiers of Computer Science, refType=null, unstructuredReference=DONG X, YU Z, CAO W, et al. A survey on ensemble learning[J]. Frontiers of Computer Science, 2020, 14(2): 241-258., articleTitle=A survey on ensemble learning, refAbstract=null), Reference(id=1177619626340004145, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2018, volume=19, issue=S3, pageStart=71, pageEnd=null, url=null, language=null, rfNumber=[10], rfOrder=15, authorNames=KEVIN M, CARLOS ND, RAZIUR R, journalName=BMC Bioinformatics, refType=null, unstructuredReference=KEVIN M, CARLOS ND, RAZIUR R, et al. Investigation of model stacking for drug sensitivity prediction[J]. BMC Bioinformatics, 2018, 19(S3): 71., articleTitle=Investigation of model stacking for drug sensitivity prediction, refAbstract=null), Reference(id=1177619626428084531, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2009, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[11], rfOrder=16, authorNames=SU L, LIAO H, YU Z, journalName=Ensemble learning for question classification, refType=null, unstructuredReference=SU L, LIAO H, YU Z, et al. Ensemble learning for question classification[Z]. 2009., articleTitle=null, refAbstract=null), Reference(id=1177619626549719349, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=197, issue=null, pageStart=106997, pageEnd=null, url=null, language=null, rfNumber=[12], rfOrder=17, authorNames=XU C, DING J, QIAO Y, journalName=Computers and Electronics in Agriculture, refType=null, unstructuredReference=XU C, DING J, QIAO Y, et al. Tomato disease and pest diagnosis method based on the Stacking of prescription data[J]. Computers and Electronics in Agriculture, 2022, 197: 106997., articleTitle=Tomato disease and pest diagnosis method based on the Stacking of prescription data, refAbstract=null), Reference(id=1177619626633605431, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=18, authorNames=吕振哲, journalName=基于重采样方法的不平衡数据分类研究, refType=null, unstructuredReference=吕振哲. 基于重采样方法的不平衡数据分类研究[D]. 烟台: 烟台大学, 2024., articleTitle=null, refAbstract=null), Reference(id=1177619626713297209, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[13], rfOrder=19, authorNames=LV ZZ, journalName=Research on imbalanced data classification based on resampling methods, refType=null, unstructuredReference=LV ZZ. Research on imbalanced data classification based on resampling methods[D]. Yantai: Yantai University, 2024., articleTitle=null, refAbstract=null), Reference(id=1177619626792988987, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=115, issue=null, pageStart=108288, pageEnd=null, url=null, language=null, rfNumber=[14], rfOrder=20, authorNames=ISLAM A, BELHAOUARI SB, REHMAN AU, journalName=Applied Soft Computing, refType=null, unstructuredReference=ISLAM A, BELHAOUARI SB, REHMAN AU, et al. Knnor: An oversampling technique for imbalanced datasets[J]. Applied Soft Computing, 2022, 115: 108288., articleTitle=Knnor: An oversampling technique for imbalanced datasets, refAbstract=null), Reference(id=1177619626876875069, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=252, issue=null, pageStart=124118, pageEnd=null, url=null, language=null, rfNumber=[15], rfOrder=21, authorNames=BELHAOUARI SB, ISLAM A, KASSOUL K, journalName=Expert Systems with Applications, refType=null, unstructuredReference=BELHAOUARI SB, ISLAM A, KASSOUL K, et al. Oversampling techniques for imbalanced data in regression[J]. Expert Systems with Applications, 2024, 252: 124118., articleTitle=Oversampling techniques for imbalanced data in regression, refAbstract=null), Reference(id=1177619626960761151, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=39, issue=6, pageStart=1615, pageEnd=1621, url=null, language=null, rfNumber=[16], rfOrder=22, authorNames=周玉, 孙红玉, 房倩, journalName=计算机应用研究, refType=null, unstructuredReference=周玉, 孙红玉, 房倩, 等. 不平衡数据集分类方法研究综述[J]. 计算机应用研究, 2022, 39(6): 1615-1621., articleTitle=不平衡数据集分类方法研究综述, refAbstract=null), Reference(id=1177619627044647233, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=39, issue=6, pageStart=1615, pageEnd=1621, url=null, language=null, rfNumber=[16], rfOrder=23, authorNames=ZHOU Y, SUN HY, FANG Q, journalName=Application Research of Computer, refType=null, unstructuredReference=ZHOU Y, SUN HY, FANG Q, et al. Review of imbalanced data classification methods[J]. Application Research of Computer, 2022, 39(6): 1615-1621., articleTitle=Review of imbalanced data classification methods, refAbstract=null), Reference(id=1177619627170476356, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2021, volume=9, issue=null, pageStart=109960, pageEnd=109975, url=null, language=null, rfNumber=[17], rfOrder=24, authorNames=KHUSHI M, SHAUKAT K, ALAM TM, journalName=IEEE Access, refType=null, unstructuredReference=KHUSHI M, SHAUKAT K, ALAM TM, et al. A comparative performance analysis of data resampling methods on imbalance medical data[J]. IEEE Access, 2021, 9: 109960-109975., articleTitle=A comparative performance analysis of data resampling methods on imbalance medical data, refAbstract=null), Reference(id=1177619627287916868, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=136, issue=null, pageStart=108872, pageEnd=null, url=null, language=null, rfNumber=[18], rfOrder=25, authorNames=LIU N, BOUZEMBRAK Y, VANDEN-BULK LM, journalName=Food Control, refType=null, unstructuredReference=LIU N, BOUZEMBRAK Y, VANDEN-BULK LM, et al. Automated food safety early warning system in the dairy supply chain using machine learning[J]. Food Control, 2022, 136: 108872., articleTitle=Automated food safety early warning system in the dairy supply chain using machine learning, refAbstract=null), Reference(id=1177619627384385862, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=26, authorNames=梁玲玲, journalName=食品安全风险预警方法研究及应用, refType=null, unstructuredReference=梁玲玲. 食品安全风险预警方法研究及应用[D]. 北京: 北京化工大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619627447300422, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[19], rfOrder=27, authorNames=LIANG LL, journalName=Research and application on food safety risk early warning method, refType=null, unstructuredReference=LIANG LL. Research and application on food safety risk early warning method[D]. Beijing: Beijing University of Chemical Technology, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619627518603592, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2018, volume=27, issue=6, pageStart=64, pageEnd=70, url=null, language=null, rfNumber=[20], rfOrder=28, authorNames=蔡照鹏, 徐林, journalName=河南城建学院学报, refType=null, unstructuredReference=蔡照鹏, 徐林. 易腐食品安全预警云体系构建研究[J]. 河南城建学院学报, 2018, 27(6): 64-70., articleTitle=易腐食品安全预警云体系构建研究, refAbstract=null), Reference(id=1177619627568935242, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2018, volume=27, issue=6, pageStart=64, pageEnd=70, url=null, language=null, rfNumber=[20], rfOrder=29, authorNames=CAI ZP, XU L, journalName=Journal of Henan University of Urban Construction, refType=null, unstructuredReference=CAI ZP, XU L. Early warning method of fresh chicken preservation period based on SVM[J]. Journal of Henan University of Urban Construction, 2018, 27(6): 64-70., articleTitle=Early warning method of fresh chicken preservation period based on SVM, refAbstract=null), Reference(id=1177619627669598540, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=34, issue=11, pageStart=315, pageEnd=323, url=null, language=null, rfNumber=[21], rfOrder=30, authorNames=杨鸿雁, 田英杰, journalName=管理评论, refType=null, unstructuredReference=杨鸿雁, 田英杰. 机器学习在食品安全风险预警及抽检方案制订中的应用研究[J]. 管理评论, 2022, 34(11): 315-323., articleTitle=机器学习在食品安全风险预警及抽检方案制订中的应用研究, refAbstract=null), Reference(id=1177619627749290318, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=34, issue=11, pageStart=315, pageEnd=323, url=null, language=null, rfNumber=[21], rfOrder=31, authorNames=YANG HY, TIAN YJ, journalName=Management Review, refType=null, unstructuredReference=YANG HY, TIAN YJ. Application research of machine learning in food safety risk early warning and sampling inspection program[J]. Management Review, 2022, 34(11): 315-323., articleTitle=Application research of machine learning in food safety risk early warning and sampling inspection program, refAbstract=null), Reference(id=1177619627870925137, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=40, issue=1, pageStart=150, pageEnd=158, url=null, language=null, rfNumber=[22], rfOrder=32, authorNames=王小艺, 王姿懿, 赵峙尧, journalName=食品科学技术学报, refType=null, unstructuredReference=王小艺, 王姿懿, 赵峙尧, 等. 集成改进AHP与XGBoost算法的食品安全风险预测模型: 以大米为例[J]. 食品科学技术学报, 2022, 40(1): 150-158., articleTitle=集成改进AHP与XGBoost算法的食品安全风险预测模型: 以大米为例, refAbstract=null), Reference(id=1177619628005142866, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=40, issue=1, pageStart=150, pageEnd=158, url=null, language=null, rfNumber=[22], rfOrder=33, authorNames=WANG XY, WANG ZY, ZHAO ZY, journalName=Journal of Food Science and Technology, refType=null, unstructuredReference=WANG XY, WANG ZY, ZHAO ZY, et al. A food safety risk forecast model integrated with improved AHP and XGBoost algorithm: A case study of rice[J]. Journal of Food Science and Technology, 2022, 40(1): 150-158., articleTitle=A food safety risk forecast model integrated with improved AHP and XGBoost algorithm: A case study of rice, refAbstract=null), Reference(id=1177619628076446035, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=1, pageStart=197, pageEnd=207, url=null, language=null, rfNumber=[23], rfOrder=34, authorNames=高亚男, 王文倩, 王建新, journalName=食品科学, refType=null, unstructuredReference=高亚男, 王文倩, 王建新. 集成模糊层级划分的LightGBM食品安全风险预警模型: 以肉制品为例[J]. 食品科学, 2021, 42(1): 197-207., articleTitle=集成模糊层级划分的LightGBM食品安全风险预警模型: 以肉制品为例, refAbstract=null), Reference(id=1177619628147749204, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2021, volume=42, issue=1, pageStart=197, pageEnd=207, url=null, language=null, rfNumber=[23], rfOrder=35, authorNames=GAO YN, WANG WQ, WANG JX, journalName=Food Science, refType=null, unstructuredReference=GAO YN, WANG WQ, WANG JX. A food safety risk prewarning model using lightGBM integrated with fuzzy hierarchy partition: A case study for meat products[J]. Food Science, 2021, 42(1): 197-207., articleTitle=A food safety risk prewarning model using lightGBM integrated with fuzzy hierarchy partition: A case study for meat products, refAbstract=null), Reference(id=1177619628252606806, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=36, authorNames=卢清华, journalName=CatBoost算法在居民食品安全满意度调查中的应用—以上海市J区为例, refType=null, unstructuredReference=卢清华. CatBoost算法在居民食品安全满意度调查中的应用—以上海市J区为例[D]. 烟台: 鲁东大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619628336492888, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[24], rfOrder=37, authorNames=LU QH, journalName=Application of CatBoost algorithm in residents' food safety satisfaction survey—Take Shanghai J District as an example, refType=null, unstructuredReference=LU QH. Application of CatBoost algorithm in residents' food safety satisfaction survey—Take Shanghai J District as an example[D]. Yantai: Ludong University, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619628428767578, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2020, volume=30, issue=9, pageStart=109, pageEnd=114, url=null, language=null, rfNumber=[25], rfOrder=38, authorNames=甄俊涛, 刘臣, journalName=计算机技术与发展, refType=null, unstructuredReference=甄俊涛, 刘臣. 高维数据多标签分类的食品安全预警研究[J]. 计算机技术与发展, 2020, 30(9): 109-114., articleTitle=高维数据多标签分类的食品安全预警研究, refAbstract=null), Reference(id=1177619628521042268, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2020, volume=30, issue=9, pageStart=109, pageEnd=114, url=null, language=null, rfNumber=[25], rfOrder=39, authorNames=ZHEN JT, LIU C, journalName=Computer Technology and Development, refType=null, unstructuredReference=ZHEN JT, LIU C. Research on food safety early warning of multilabel classification of high dimensional data[J]. Computer Technology and Development, 2020, 30(9): 109-114., articleTitle=Research on food safety early warning of multilabel classification of high dimensional data, refAbstract=null), Reference(id=1177619628613316958, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=40, authorNames=唐昊, journalName=基于机器学习的典型食品安全风险预警研究, refType=null, unstructuredReference=唐昊. 基于机器学习的典型食品安全风险预警研究[D]. 北京: 北京化工大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619628734951776, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[26], rfOrder=41, authorNames=TANG H, journalName=Research on risk early warning of typical food safety based on machine learning, refType=null, unstructuredReference=TANG H. Research on risk early warning of typical food safety based on machine learning[D]. Beijing: Beijing University of Chemical Technology, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619628902723937, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=628, issue=null, pageStart=70, pageEnd=91, url=null, language=null, rfNumber=[27], rfOrder=42, authorNames=MOAZEN H, MOLAEI S, FARZINVASH L, journalName=Information Sciences, refType=null, unstructuredReference=MOAZEN H, MOLAEI S, FARZINVASH L, et al. PSO-ELPM: PSO with elite learning, enhanced parameter updating, and exponential mutation operator[J]. Information Sciences, 2023, 628: 70-91., articleTitle=PSO-ELPM: PSO with elite learning, enhanced parameter updating, and exponential mutation operator, refAbstract=null), Reference(id=1177619629015970146, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=10, pageStart=32, pageEnd=37, url=null, language=null, rfNumber=[28], rfOrder=43, authorNames=何浦, journalName=自动化与仪器仪表, refType=null, unstructuredReference=何浦. 基于粒子群算法的食品加工工艺参数优化模型构建[J]. 自动化与仪器仪表, 2022(10): 32-37., articleTitle=基于粒子群算法的食品加工工艺参数优化模型构建, refAbstract=null), Reference(id=1177619629225685347, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2022, volume=null, issue=10, pageStart=32, pageEnd=37, url=null, language=null, rfNumber=[28], rfOrder=44, authorNames=HE P, journalName=Automation & Instrumentation, refType=null, unstructuredReference=HE P. Construction of optimization model of food processing parameters based on particles warm optimization[J]. Automation & Instrumentation, 2022(10): 32-37., articleTitle=Construction of optimization model of food processing parameters based on particles warm optimization, refAbstract=null), Reference(id=1177619629355708772, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2024, volume=201, issue=null, pageStart=116252, pageEnd=null, url=null, language=null, rfNumber=[29], rfOrder=45, authorNames=ZHOU ZW, TIAN ML, LIU BX, journalName=LWT, refType=null, unstructuredReference=ZHOU ZW, TIAN ML, LIU BX, et al. Assessment of salmon sashimi processing conditions for Listeria monocytogenes cross-contamination and effectiveness of CLPSO-BP neural network model constructed for predicting microbial transfer[J]. LWT, 2024, 201: 116252., articleTitle=Assessment of salmon sashimi processing conditions for Listeria monocytogenes cross-contamination and effectiveness of CLPSO-BP neural network model constructed for predicting microbial transfer, refAbstract=null), Reference(id=1177619629494120805, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=46, authorNames=邢巍, journalName=基于人工神经网络和计算机视觉系统的烤肠多元安全品质预测模型构建, refType=null, unstructuredReference=邢巍. 基于人工神经网络和计算机视觉系统的烤肠多元安全品质预测模型构建[D]. 合肥: 合肥工业大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619629582201190, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[30], rfOrder=47, authorNames=XING W, journalName=Construction of multivariate safety quality prediction model of roast sausage based on artificial neural network and computer vision system, refType=null, unstructuredReference=XING W. Construction of multivariate safety quality prediction model of roast sausage based on artificial neural network and computer vision system[D]. Hefei: Hefei University of Technology, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619629674475879, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[31], rfOrder=48, authorNames=陈鹏飞, journalName=基于SA-PSO-BP神经网络的牛肉微波干燥含水率预测研究, refType=null, unstructuredReference=陈鹏飞. 基于SA-PSO-BP神经网络的牛肉微波干燥含水率预测研究[D]. 银川: 宁夏大学, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619629775139176, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=null, issue=null, pageStart=null, pageEnd=null, url=null, language=null, rfNumber=[31], rfOrder=49, authorNames=CHEN PF, journalName=Research on prediction of moisture content of beef microwave drying based on SA-PSO-BP neural network, refType=null, unstructuredReference=CHEN PF. Research on prediction of moisture content of beef microwave drying based on SA-PSO-BP neural network[D]. Yinchuan: Ningxia University, 2023., articleTitle=null, refAbstract=null), Reference(id=1177619629863219561, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2023, volume=230, issue=null, pageStart=120747, pageEnd=null, url=null, language=null, rfNumber=[32], rfOrder=50, authorNames=GENG Z, WANG X, JIANG Y, journalName=Expert Systems with Applications, refType=null, unstructuredReference=GENG Z, WANG X, JIANG Y, et al. Novel IAPSO-LSTM neural network for risk analysis and early warning of food safety[J]. Expert Systems with Applications, 2023, 230: 120747., articleTitle=Novel IAPSO-LSTM neural network for risk analysis and early warning of food safety, refAbstract=null), Reference(id=1177619629972271466, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2021, volume=12, issue=3, pageStart=1015, pageEnd=1020, url=null, language=null, rfNumber=[33], rfOrder=51, authorNames=宁霄, 贺鑫鑫, 金绍明, journalName=食品安全质量检测学报, refType=null, unstructuredReference=宁霄, 贺鑫鑫, 金绍明, 等. 食品快速检测方法的评价及其贝叶斯推断[J]. 食品安全质量检测学报, 2021, 12(3): 1015-1020., articleTitle=食品快速检测方法的评价及其贝叶斯推断, refAbstract=null), Reference(id=1177619630135849323, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, doi=null, pmid=null, pmcid=null, year=2021, volume=12, issue=3, pageStart=1015, pageEnd=1020, url=null, language=null, rfNumber=[33], rfOrder=52, authorNames=NING X, HE XX, JIN SM, journalName=Journal of Food Safety and Quality, refType=null, unstructuredReference=NING X, HE XX, JIN SM, et al. Evaluation of food rapid detection methods and their Bayesian inference[J]. Journal of Food Safety and Quality, 2021, 12(3): 1015-1020., articleTitle=Evaluation of food rapid detection methods and their Bayesian inference, refAbstract=null)], funds=[Fund(id=1177619624326738196, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, awardId=2024-50, language=CN, fundingSource=2024年上海市市场监督管理局科技项目(2024-50), fundOrder=null, country=null), Fund(id=1177619624414818582, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, awardId=2022MK034, language=CN, fundingSource=国家市场监督管理总局科技计划项目(2022MK034), fundOrder=null, country=null)], companyList=[AuthorCompany(id=1177619618744119440, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, xref=null, ext=[AuthorCompanyExt(id=1177619618752508048, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China), AuthorCompanyExt(id=1177619618760896657, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, companyId=1177619618744119440, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=上海市市场监督管理局信息应用研究中心, 上海 200030)])], figs=[ArticleFig(id=1177619621831127267, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Fig.1, caption=Stacking model procedure flowchart, figureFileSmall=t2sNG/NQZiUy3LfE8VRH8A==, figureFileBig=OfIPT+YVq+rB1Hyiko5Gkw==, tableContent=null), ArticleFig(id=1177619621931790567, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=图1, caption=Stacking模型程序流程图, figureFileSmall=t2sNG/NQZiUy3LfE8VRH8A==, figureFileBig=OfIPT+YVq+rB1Hyiko5Gkw==, tableContent=null), ArticleFig(id=1177619622086979817, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Fig.2, caption=AUC values of the prediction results of different machine learning models based on sampling results of fenthion in vegetables, figureFileSmall=T7ZecV5O4AqMpqsNomvd2A==, figureFileBig=zDyAiVFDsybOYh2b1ropuQ==, tableContent=null), ArticleFig(id=1177619622183448812, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=图2, caption=基于蔬菜中倍硫磷抽检结果的不同机器学习模型预测结果的AUC值, figureFileSmall=T7ZecV5O4AqMpqsNomvd2A==, figureFileBig=zDyAiVFDsybOYh2b1ropuQ==, tableContent=null), ArticleFig(id=1177619622330249455, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Fig.3, caption=Confusion matrix (A) and PR curve (B) for PSO-stacking model predictions, figureFileSmall=9oJ3lL06I8xNIU7ty00KQw==, figureFileBig=k8r4MFq1uT1Yc3MJF/s+DA==, tableContent=null), ArticleFig(id=1177619622435107058, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=图3, caption=PSO-Stacking模型预测结果的混淆矩阵(A)与PR曲线(B), figureFileSmall=9oJ3lL06I8xNIU7ty00KQw==, figureFileBig=k8r4MFq1uT1Yc3MJF/s+DA==, tableContent=null), ArticleFig(id=1177619622594490613, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Fig.4, caption=Comparison of ROC curves for different machine learning prediction models, figureFileSmall=7pR+YYKAuoVA4xuwrs7sig==, figureFileBig=8XT8iu2KMQllDPKoew/B+w==, tableContent=null), ArticleFig(id=1177619622699348214, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=图4, caption=不同机器学习预测模型的ROC曲线对比图, figureFileSmall=7pR+YYKAuoVA4xuwrs7sig==, figureFileBig=8XT8iu2KMQllDPKoew/B+w==, tableContent=null), ArticleFig(id=1177619622888091897, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Fig.5, caption=Heat map of feature distribution under SHAP-based sample clustering, figureFileSmall=7F2qiJm7EkPFoFgrjEnGWA==, figureFileBig=zFxL6T2YuoUw5+8E+es6tA==, tableContent=null), ArticleFig(id=1177619622980366587, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=图5, caption=基于SHAP的样本聚类下特征分布热力图

注: 红色块代表正SHAP值, 表示特征对预测结果有显著的正面贡献, 蓝色块代表负SHAP值, 表明特征的增加会导致预测结果的降低。

, figureFileSmall=7F2qiJm7EkPFoFgrjEnGWA==, figureFileBig=zFxL6T2YuoUw5+8E+es6tA==, tableContent=null), ArticleFig(id=1177619623106195709, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Fig.6, caption=Characteristic influence diagrams of micro sample, figureFileSmall=MSe+qNqZhWw3P2nK7/Qsfg==, figureFileBig=DN02k6HjycFmF9hUWQFiUA==, tableContent=null), ArticleFig(id=1177619623244607742, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=图6, caption=微观样本特征影响图

注: 红色标签代表特征对预测结果产生正向影响, 蓝色代表负向影响, 长短代表影响大小, 标签下方是该特征对应的子特征检序列。A. 第100个样本的预测结果; B. 第394个样本的预测结果; C. 第1625个样本的预测结果; 所有样本均随机抽取。

, figureFileSmall=MSe+qNqZhWw3P2nK7/Qsfg==, figureFileBig=DN02k6HjycFmF9hUWQFiUA==, tableContent=null), ArticleFig(id=1177619623315910911, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Table 1, caption=

Sampling results of fenthion in vegetables available for sale in Shanghai

, figureFileSmall=null, figureFileBig=null, tableContent=
蔬菜品种 总件数 不合格件数 不合格率/%
豆类蔬菜 2329 53 2.3
茄果类蔬菜 868 2 0.2
瓜类蔬菜 623 0 0.0
叶菜类蔬菜 46 0 0.0
鳞茎类蔬菜 15 0 0.0
水生类蔬菜 4 0 0.0
根茎类和薯芋类蔬菜 4 0 0.0
合计 3889 55 1.4
), ArticleFig(id=1177619623403991297, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=表1, caption=

上海市市售蔬菜中倍硫磷的抽检结果

, figureFileSmall=null, figureFileBig=null, tableContent=
蔬菜品种 总件数 不合格件数 不合格率/%
豆类蔬菜 2329 53 2.3
茄果类蔬菜 868 2 0.2
瓜类蔬菜 623 0 0.0
叶菜类蔬菜 46 0 0.0
鳞茎类蔬菜 15 0 0.0
水生类蔬菜 4 0 0.0
根茎类和薯芋类蔬菜 4 0 0.0
合计 3889 55 1.4
), ArticleFig(id=1177619623496265987, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Table 2, caption=

Sampling results for fenthion in vegetables collected during different months

, figureFileSmall=null, figureFileBig=null, tableContent=
抽样月份 总件数 不合格件数 不合格率/%
1 40 0 0.0
2 95 12 12.6
3 117 9 7.7
4 198 4 2.0
5 240 2 0.8
6 334 0 0.0
7 592 0 0.0
8 708 3 0.4
9 649 11 1.7
10 660 9 1.4
11 243 5 2.1
12 13 0 0.0
), ArticleFig(id=1177619623592734982, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=表2, caption=

不同抽样月份蔬菜中倍硫磷的抽检结果

, figureFileSmall=null, figureFileBig=null, tableContent=
抽样月份 总件数 不合格件数 不合格率/%
1 40 0 0.0
2 95 12 12.6
3 117 9 7.7
4 198 4 2.0
5 240 2 0.8
6 334 0 0.0
7 592 0 0.0
8 708 3 0.4
9 649 11 1.7
10 660 9 1.4
11 243 5 2.1
12 13 0 0.0
), ArticleFig(id=1177619623680815367, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Table 3, caption=

Sampling results of fenthion in vegetables at different sampling locations

, figureFileSmall=null, figureFileBig=null, tableContent=
抽样环节 抽样场所 总件数/件 不合格件数/件 不合格率/%
流通环节 农贸市场 1347 12 0.9
超市卖场 754 14 1.9
网络购物 557 16 2.9
食品商店 425 8 1.9
其他场所 247 0 0.0
合计 3330 50 1.5
餐饮环节 单位食堂 260 2 0.8
中型餐馆 128 1 0.8
大型餐馆 90 0 0.0
小型餐馆 81 2 2.5
合计 559 5 0.9
), ArticleFig(id=1177619623773090057, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=表3, caption=

不同抽样场所蔬菜中倍硫磷的抽检结果

, figureFileSmall=null, figureFileBig=null, tableContent=
抽样环节 抽样场所 总件数/件 不合格件数/件 不合格率/%
流通环节 农贸市场 1347 12 0.9
超市卖场 754 14 1.9
网络购物 557 16 2.9
食品商店 425 8 1.9
其他场所 247 0 0.0
合计 3330 50 1.5
餐饮环节 单位食堂 260 2 0.8
中型餐馆 128 1 0.8
大型餐馆 90 0 0.0
小型餐馆 81 2 2.5
合计 559 5 0.9
), ArticleFig(id=1177619623844393227, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Table 4, caption=

Optimisation results based on PSO algorithm base model parameters

, figureFileSmall=null, figureFileBig=null, tableContent=
基模型 树的数量 最大深度 学习率
RF 233 6 -
GB 968 5 0.001
XGBoost 763 5 0.47
LGBM 767 6 0.23
CatBoost 393 0.83 0.47
), ArticleFig(id=1177619623961833741, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=表4, caption=

基于PSO算法基模型参数寻优结果

, figureFileSmall=null, figureFileBig=null, tableContent=
基模型 树的数量 最大深度 学习率
RF 233 6 -
GB 968 5 0.001
XGBoost 763 5 0.47
LGBM 767 6 0.23
CatBoost 393 0.83 0.47
), ArticleFig(id=1177619624045719823, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=EN, label=Table 5, caption=

Comparison of different machine learning models for predicting sampling results of fenthion in vegetables

, figureFileSmall=null, figureFileBig=null, tableContent=
基模型 精确率 召回率 F1值 AUC值
PSO-RF 0.43 0.83 0.57 0.86
PSO-GB 0.45 0.83 0.59 0.86
PSO-XGBoost 0.69 0.75 0.72 0.86
PSO-LGBM 0.71 0.83 0.77 0.90
PSO-CatBoost 0.83 0.83 0.83 0.91
PSO-Stacking 0.91 0.83 0.87 0.91
), ArticleFig(id=1177619624121217297, tenantId=1146029695717560320, journalId=1149652044408987649, articleId=1153429494946844900, language=CN, label=表5, caption=

不同机器学习模型对蔬菜中倍硫磷抽检结果预测对比

, figureFileSmall=null, figureFileBig=null, tableContent=
基模型 精确率 召回率 F1值 AUC值
PSO-RF 0.43 0.83 0.57 0.86
PSO-GB 0.45 0.83 0.59 0.86
PSO-XGBoost 0.69 0.75 0.72 0.86
PSO-LGBM 0.71 0.83 0.77 0.90
PSO-CatBoost 0.83 0.83 0.83 0.91
PSO-Stacking 0.91 0.83 0.87 0.91
)], attaches=null, journal=Journal(id=1146438185598431260, delFlag=0, nameCn=食品安全质量检测学报, nameEn=Journal of Food Safety & Quality, nameHistory1=null, nameHistory2=null, issn=2095-0381, eissn=null, cn=11-5956/TS, coden=null, periodic=3, language=CN, oaType=0, ccby=CC BY-NC-ND, 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=4kwA5ri4VfzavOn19fwc2g==, journalPrice=null, startedYear=null, abbrevIsoEn=J Food Saf Qual, journalRemark=null, publicationField=null, createdTime=1751261763241, updatedTime=1754445151803, createdBy=18614031015, updatedBy=13701087609, firstLetterCn=J, firstLetterEn=J, subjectCode=Engineering, subjectName=工程, subjectCodeEn=Engineering, subjectNameEn=null, picCn=4kwA5ri4VfzavOn19fwc2g==, picEn=eHhRQNEus+t1f6yOGnBJ3w==, jcr=null, cjcr=null, exts=[JournalExt(id=1159790285120618579, language=CN, name=食品安全质量检测学报, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1754445151834, updatedTime=1754445151834, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://chinafoodj.ijournals.cn/ch/first_menu.aspx?parent_id=20160111022419868, submissionAuthorUrl=http://chinafoodj.ijournals.cn/ch/author/login.aspx, submissionEditorUrl=http://chinafoodj.ijournals.cn/ch/login.aspx, submissionReviewUrl=http://chinafoodj.ijournals.cn/ch/auditor/login.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""}), JournalExt(id=1159790285187727444, language=EN, name=Journal of Food Safety & Quality, nameHistory1=null, nameHistory2=null, managedBy=, sponsoredBy=, publishedBy=, editorOffice=, officeProv=null, officeCity=null, officeAddr=, officeZip=, editDirector=null, officeDirector=null, officePhone=null, coverPicUrl=null, journalRemark=, submitArticleUrl=null, websiteUrl=, createdTime=1754445151850, updatedTime=1754445151850, createdBy=13701087609, updatedBy=13701087609, submissionGuidelinesUrl=http://chinafoodj.ijournals.cn/ch/first_menu.aspx?parent_id=20160111022419868, submissionAuthorUrl=http://chinafoodj.ijournals.cn/ch/author/login.aspx, submissionEditorUrl=http://chinafoodj.ijournals.cn/ch/login.aspx, submissionReviewUrl=http://chinafoodj.ijournals.cn/ch/auditor/login.aspx, submissionCeEditorUrl=, submissionAeEditorUrl=, option={"copyright":""})], databaseList=null, tenantJournalId=1149652044408987649, websiteList=[Website(id=1151872930754474249, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1149652044408987649, 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/spaq/CN, language=CN, createTime=1752557507456, createBy=18614031015, updateTime=1752558523388, updateBy=18614031015, name=食品安全质量检测学报, tplId=1146099689490845704, title=食品安全质量检测学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1151887749801407249, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930754474249, code=articleTextType, value=kx, createTime=1752561040592, updateTime=1752561040592, creator=18614031015, updator=18614031015), WebsiteProps(id=1151887749776241422, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930754474249, code=banner, value=null, createTime=1752561040586, updateTime=1752561040586, creator=18614031015, updator=18614031015), WebsiteProps(id=1151887749767852813, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930754474249, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic?fileId=RhDdm4lBDQfNHl8cX0659g==, createTime=1752561040584, updateTime=1752561040584, creator=18614031015, updator=18614031015), WebsiteProps(id=1151887749793018640, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930754474249, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/CN/file/pic, createTime=1752561040590, updateTime=1752561040590, creator=18614031015, updator=18614031015), WebsiteProps(id=1151887749784630031, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930754474249, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_cn_619/, createTime=1752561040588, updateTime=1752561040588, creator=18614031015, updator=18614031015)]), Website(id=1151872930855137548, webName=null, webTitle=null, webDomain=null, webCopyrigh=null, webIpcNo=null, seoTitle=null, seoKeywords=null, seoDescription=null, tenantJournalId=null, journalId=1149652044408987649, 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/spaq/EN, language=EN, createTime=1752557507480, createBy=18614031015, updateTime=1752558528290, updateBy=18614031015, name=食品安全质量检测学报, tplId=1146101810881728533, title=食品安全质量检测学报, delFlag=0, indexPage=/home, props=[WebsiteProps(id=1155904094024884374, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930855137548, code=articleTextType, value=kx, createTime=1753518611675, updateTime=1753518611675, creator=18614031015, updator=18614031015), WebsiteProps(id=1155904094008107155, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930855137548, code=banner, value=null, createTime=1753518611671, updateTime=1753518611671, creator=18614031015, updator=18614031015), WebsiteProps(id=1155904094003912850, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930855137548, code=logo, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic?fileId=RhDdm4lBDQfNHl8cX0659g==, createTime=1753518611670, updateTime=1753518611670, creator=18614031015, updator=18614031015), WebsiteProps(id=1155904094020690069, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930855137548, code=picServerUrl, value=https://castjournals.cast.org.cn/joweb/kjdb/EN/file/pic, createTime=1753518611674, updateTime=1753518611674, creator=18614031015, updator=18614031015), WebsiteProps(id=1155904094016495764, tenantId=1146029695717560320, journalId=null, journalGroupId=null, siteId=1151872930855137548, code=staticResourcePath, value=https://castjournals.cast.org.cn/joweb/cast_kjdb_en_623/, createTime=1753518611673, updateTime=1753518611673, creator=18614031015, updator=18614031015)])], journalTitle=食品安全质量检测学报, weixinUrl=null, journalUrl=null, iacademicId=null, status=0, seqNo=null, journalTitleEn=Journal of Food Safety & Quality, journalPhotoCn=4kwA5ri4VfzavOn19fwc2g==, journalPhotoEn=eHhRQNEus+t1f6yOGnBJ3w==, 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=null, provinceCode=null, provinceName=null, collectFlag=false), detailUrlCn=https://castjournals.cast.org.cn/joweb/spaq/CN/10.19812/j.cnki.jfsq11-5956/ts.20241008004, detailUrlEn=https://castjournals.cast.org.cn/joweb/spaq/EN/10.19812/j.cnki.jfsq11-5956/ts.20241008004, pdfUrlCn=https://castjournals.cast.org.cn/joweb/spaq/CN/PDF/10.19812/j.cnki.jfsq11-5956/ts.20241008004, pdfUrlEn=https://castjournals.cast.org.cn/joweb/spaq/EN/PDF/10.19812/j.cnki.jfsq11-5956/ts.20241008004, aliStartDate=null, aliEndDate=null, collectionFlag=false, citedCount=null, citedUrl=null, reference=null)
收藏切换
基于粒子群优化集成学习算法堆叠模型预测蔬菜中倍硫磷的抽检结果
收藏切换
PDF下载
周子文 , 范志仪 , 彭少杰 *
食品安全质量检测学报 | 本期专题:农兽药残留研究与检测 2025,16(5): 187-196
收起
收藏切换
食品安全质量检测学报 | 本期专题:农兽药残留研究与检测 2025, 16(5): 187-196
基于粒子群优化集成学习算法堆叠模型预测蔬菜中倍硫磷的抽检结果
全屏
周子文 , 范志仪, 彭少杰*
作者信息
  • 上海市市场监督管理局信息应用研究中心, 上海 200030
  • 周子文(1999—), 男, 硕士, 主要研究方向为食品质量与安全(风险监测与评估)。E-mail:

通讯作者:

* 彭少杰(1968—), 男, 硕士, 主任医师, 主要研究方向为食品安全风险监测与评估。E-mail:
Stacked generalization model prediction of fenthion spot-check results in vegetables based on particle swarm optimization ensemble learning algorithm
Zi-Wen ZHOU , Zhi-Yi FAN, Shao-Jie PENG*
Affiliations
  • Information Application Research Center of Shanghai Municipal Administration for Market Regulation, Shanghai 200030, China
出版时间: 2025-03-15 doi: 10.19812/j.cnki.jfsq11-5956/ts.20241008004
文章导航
收藏切换

目的 建立基于粒子群(particle swarm optimization, PSO)算法优化堆叠模型(stacked generalization, Stacking)的蔬菜安全风险预测模型, 对上海市市售蔬菜中倍硫磷的抽检结果进行预测。方法 基于2021—2023年上海市市售蔬菜中倍硫磷的抽检数据, 选取任务类型、抽样地区、抽样环节、抽样场所、抽样月份、检测机构、蔬菜品种作为特征变量, 以蔬菜中倍硫磷的抽检结果是否合格为目标变量; 采用十折交叉验证筛选优良机器学习模型、重采样方法, 经PSO算法优化模型参数后构建PSO-Stacking预测模型。结果 3889件蔬菜中检出倍硫磷阳性样品55件, 不合格率为1.4%。其中, 豆类蔬菜不合格率最高(2.3%), 其次为茄果类(0.2%)。筛选得到基模型, 包括随机森林(random forest, RF)、类别特征梯度提升树(categorical boosting, CatBoost)、梯度提升(gradient boosting, GB)、极端梯度提升(extreme gradient boosting, XGBoost)和轻量级梯度提升机(light gradient boosting machine, LGBM), 最佳重采样方法为自适应合成抽样(adaptive synthetic sampling, ADASYN)技术。PSO-Stacking模型在测试集上的精确率(0.91)、召回率(0.83)、F1值(0.87)和曲线下面积(area under the curve, AUC)值(0.91)均为最高。结论 PSO-Stacking模型在不均衡食品安全抽检数据中表现优异, 能准确预测蔬菜中倍硫磷不合格样本, 为蔬菜监督抽检及风险预警提供技术支撑。

蔬菜  /  倍硫磷  /  粒子群算法  /  堆叠模型  /  机器学习  /  食品安全

Objective To establish a vegetable safety risk prediction model based on the particle swarm optimization (PSO) algorithm and the stacked generalization (Stacking) model, and to predict the sampling results of fenthion in vegetables sold in Shanghai. Methods Based on the sampling data of fenthion in vegetables sold in Shanghai from 2021 to 2023, task type, sampling area, sampling link, sampling place, sampling month, testing institution, and vegetable variety were selected as feature variables. The target variable was whether the sampling results for fenthion in vegetables were qualified. The PSO-Stacking prediction model was constructed using ten-fold cross-validation to select effective machine learning models and resampling methods and optimized the model parameters using the PSO algorithm. Results Fenthion-positive samples were found in 55 out of 3889 vegetable samples, with an overall failure rate of 1.4%. Bean vegetables had the highest rate at 2.3%, followed by eggplant and fruiting vegetables at 0.2%. The base models were obtained through screening, including Random Forest (RF), categorical boosting (CatBoost), gradient boosting (GB), extreme gradient Boosting (XGBoost), and light gradient boosting machine (LGBM). The best resampling technique was adaptive synthetic sampling (ADASYN). The PSO-Stacking model achieved the highest precision (0.91), recall (0.83), F1 score (0.87), and area under the curve (AUC) value (0.91) on the test set. Conclusion The PSO-Stacking model effectively addresses imbalanced food safety sampling data, accurately predicts the unqualified fenthion samples in vegetables, and provides technical support for vegetable supervision, sampling and risk warning.

vegetables  /  fenthion  /  particle swarm algorithm  /  stacked generalization model  /  machine learning  /  food safety
周子文, 范志仪, 彭少杰. 基于粒子群优化集成学习算法堆叠模型预测蔬菜中倍硫磷的抽检结果. 食品安全质量检测学报, 2025 , 16 (5) : 187 -196 . DOI: 10.19812/j.cnki.jfsq11-5956/ts.20241008004
Zi-Wen ZHOU, Zhi-Yi FAN, Shao-Jie PENG. Stacked generalization model prediction of fenthion spot-check results in vegetables based on particle swarm optimization ensemble learning algorithm[J]. Journal of Food Safety & Quality, 2025 , 16 (5) : 187 -196 . DOI: 10.19812/j.cnki.jfsq11-5956/ts.20241008004
有机磷农药是一种含有有机磷酸盐的有机化合物, 广泛应用于农业, 以保护作物免受害虫侵害, 提高农产品产量并减少损失。然而, 不合理使用可能导致环境污染, 潜在危害环境和人体健康[1]。据相关统计[2-3], 每年因农药中毒就医的人群中急性有机磷农药占比约20%~50%, 中毒病死率约3%~58.6%。倍硫磷是有机磷农药的一种, 与其他杀虫剂相比毒性低、效果好, 在蔬菜中广泛使用, 但其具有神经毒性[4], 会在人体和动物体内蓄积, 因此必须严格控制蔬菜中倍硫磷的残留量。
食品安全预警监测有助于在事前发现食品风险隐患, 变事后被动监管为事前主动监管保障, 以达到风险预警和保障食品安全的目的[5]。食品安全抽检数据普遍存在不均衡性, 大多数结果为合格, 不合格仅占少数。若直接使用原始数据建模, 模型可能仅预测所有样本为合格, 表面性能虽佳, 但实际无效。为了解决这个问题, WANG等[6]建立了加权贝叶斯网络, 用于食品中重金属阳性的预测。LI等[7]通过切比雪夫距离计算的高斯分布重采样方法, 有效地解决乳制品抽检数据不均衡的问题。吴帅[5]针对食品安全抽检不均衡数据集, 提出了一种基于密度峰值和K近邻的密度均衡采样算法, 构建了代价敏感和粒子群改进的反向传播(backpropagation, BP)神经网络模型。
机器学习具备高效的分析与处理数据的能力, 能够准确识别食品安全风险中与风险发生相关联的多维度特征[8]。集成模型是应对许多机器学习挑战的先进解决方案, 利用机器学习模型的多样性来提高泛化性能和稳定性, 在分类问题上表现出巨大的优势[9-10]。机器学习中最具代表性的集成模型是引导聚集算法(bootstrap aggregating, Bagging)、提升方法(boosting)和堆叠模型(stacked generalization, Stacking)[11]。其中, Stacking模型有两层结构, 由多个不同基模型和一个元模型组成, 在分类问题上表现良好, 特别是在不均衡数据分类中[12]。重采样方法在构建不均衡数据集的预测模型中非常重要。研究者常通过数据重采样的方法来应对不均衡数据集带来的问题[13]。数据重采样方法包括过采样、欠采样以及混合采样方法: 过采样方法主要有合成少数类过采样(synthetic minority over-sampling technique, SMOTE)技术、基于边界改良合成少数类过采样(Borderline-SMOTE)技术、自适应合成抽样(adaptive synthetic sampling, ADASYN)技术以及随机过采样(random over sampler, ROS)技术等[14-15]; 欠采样方法包括聚类质心(Cluster-Centroid)技术、随机欠采样(random under sampler, RUS)技术等[16]; 常见的混合采样方法包括合成少数类过采样技术与编辑最近邻方法(synthetic minority over-sampling technique with edited nearest neighbours, SMOTEENN)和合成少数类过采样技术与托梅克链接(synthetic minority over-sampling technique with tomek links, SMOTETomek)等[17]。过采样技术能有效的缓解类别不平衡的问题, 但也会导致模型的泛化能力与预测性能降低, 使模型的广泛适用性不足。
目前用于食品安全风险分析和预警机器学习模型主要包括贝叶斯网络(Bayesian network, BN)[18]、支持向量机(support vector machine, SVM)[19-20]、随机森林(random forest, RF)[21]、多层感知机(multilayer perceptron, MLP)、梯度提升(gradient boosting, GB)算法、极端梯度提升树算法(extreme gradient boosting, XGBoost)[22]、轻量级梯度提升机(light gradient boosting machine, LGBM)[23]、基于对称决策树算法的GBDT框架(categorical boosting, CatBoost)[24]、人工神经网络(artificial neural network, ANN)[25-26]等。
粒子群算法(particle swarm optimization, PSO)因其收敛速度快、参数少、易实现的特点, 针对高维优化问题能够短时间内地收敛于最优解[27]。PSO算法使用场景广泛, 目前在食品领域的主要应用以优化BP神经网络的超参数为主[28-29]。邢巍[30]构建了PSO-BP神经网络, 并结合计算机视觉系统预测烤肠的安全品质。陈鹏飞[31]开发了模拟退火-PSO-BP神经网络模型, 通过模型预测肉干制品微波干燥过程产品水分含量的变化, 为食品的贮藏提供了理论依据。吴帅[5]针对不均衡的食品安全抽检数据提出了代价敏感和粒子群改进的BP神经网络模型来预警食品安全风险。GENG[32]提出了一种基于自适应粒子群优化算法的长短期记忆神经网络用于食品安全风险分析和预警。
鉴于此, 本研究基于2021—2023年上海市市售蔬菜中倍硫磷的抽检数据, 筛选出几种表现优异的机器学习模型作为初始基模型, 并通过粒子群算法优化, 构建PSO-Stacking模型。模型以任务类型、抽样地区、抽样环节、抽样场所、抽样月份、检测机构、蔬菜品种7个属性为特征变量, 预测蔬菜中倍硫磷的抽检结果, 为监管部门进一步做好基于问题导向的蔬菜抽检检测和科学监管提供技术参考。
蔬菜数据集来源于上海市市场监管部门于2021—2023年组织开展的食品安全抽检, 所有样品均随机采集于上海市的流通环节, 共抽检3889件样品。按照抽检任务类型分为监督抽检和评价性抽检; 按照蔬菜的品种分为茄果类蔬菜、瓜类蔬菜、豆类蔬菜、水生类蔬菜、叶菜类蔬菜、鳞茎类蔬菜、根茎类和薯芋类蔬菜; 按照样品采集的场所分为流通环节(包括超市、网店、农贸市场、批发市场、食品店等)和餐饮环节(包括特大型餐馆、大型餐馆、中型餐馆、小型餐馆、单位食堂等); 抽样地区覆盖了上海市的16个区。
蔬菜中倍硫磷的检验方法为GB 23200.113—2018《食品安全国家标准 植物源性食品中208种农药及其代谢物残留量的测定 气相色谱-质谱联用法》、GB 23200.121—2021《食品安全国家标准 植物源性食品中331种农药及其代谢物残留量的测定 液相色谱-质谱联用法》、GB/T 20769—2008《水果和蔬菜中450种农药及相关化学品残留量的测定 液相色谱-串联质谱法》和GB 23200.8—2016《食品安全国家标准水果和蔬菜中500种农药及相关化学品残留量的测定 气相色谱-质谱法》。本研究按照GB 2763—2021《食品安全国家标准 食品中农药最大残留限量》中蔬菜的倍硫磷最大残留限量(0.05 mg/kg)对样品的检测结果进行评价。
针对蔬菜倍硫磷原始数据, 采用十折交叉验证方法优选机器学习模型与数据重采样方法。考虑到原始数据为类别不均衡数据集, 准确率不能作为模型性能的评判指标, 因此, 本研究以曲线下面积(area under the curve, AUC)值作为十折交叉验证的目标函数, 即模型筛选的指标。
PSO将多维数据映射到速度与位置的二维空间中, 以目标函数作为适应值函数, 通过粒子最优位置(pbest)和群体最优位置(gbest)两个关键参数寻找最优解, 无需考虑参数的个数。粒子的速度和位置分别根据公式(1)进行更新:
$\left\{\begin{array}{l} v_{i d}^{k+1}=\omega v_{i d}^{k}+c_{1} r_{1}\left(p_{i d, \text { pbest }}^{k}-x_{i d}^{k}\right)+c_{2} r_{2}\left(p_{d, \text { gbest }}^{k}-x_{i d}^{k}\right) \\ x_{i d}^{k+1}=x_{i d}^{k}+v_{i d}^{k+1} \end{array}\right.$
式中, k为迭代次数; ω为惯性权重, 其取值一般随着迭代次数的增加从0.9线性递减到0.4; $x_{i d}^{k+1}, v_{i d}^{k+1}$为第k+1代粒子当前位置和速度; $x_{i d}^{k} 、 v_{i d}^{k}$为第k代粒子当前位置和速度; $p_{i d, \text { pbest }}^{k}$为第k代个体最优粒子当前位置; $p_{d, \text { gbest }}^{k}$为第k代全局最优粒子所处位置。
Stacking模型的程序流程图如图1所示, 流程如下: 首先对原始数据进行清洗、探索性数据分析与特征工程得到特征变量与目标变量; 然后通过十折交叉验证筛选机器学习模型以及最佳数据重采样方法; 随后使用粒子群算法优化初始基模型, 构建PSO-Stacking模型, 训练模型并在测试集上评估Stacking模型的整体性能。
食品安全抽检结果的原始数据包含大量属性变量, 但部分样本信息模糊且难以溯源, 导致许多字段缺失, 且存在高度共线性(如抽样地区和单位)。因此, 在特征工程的数据清洗阶段, 使用Pandas库剔除异常值和空值。结合专业分析, 并考虑地理位置、抽样场所、卫生管理和季节变化等因素, 本研究筛选出7个显著影响抽检合格率的特征: 任务类型、抽样地区、抽样环节、抽样场所、抽样月份、检测机构、蔬菜品种。其中, 不同任务类型和检测机构可能导致标准差异, 抽样地区和场所的卫生状况直接影响样本质量, 季节变化也会影响农产品安全, 而不同蔬菜品种则反映出不同的生长环境和风险特征。特征选择后, 使用类别编码将分类数据转换为数值格式, 并进行数据划分。本研究旨在科学识别和量化这些特征对抽检合格率的影响, 为食品安全监管提供决策支持。
正例和负例是机器学习和统计学中的基本概念, 用于评估模型的性能。‌在机器学习模型中正例是指所关注的识别目标, 而负例则是正例以外的其他数据。在特征工程中, 由机器学习库中的标签编码模块自动为类别变量赋值。‌本研究旨在建立能尽可能识别蔬菜倍硫磷抽检结果中的少数类不合格样本的模型, 因此, 以少数类样本(不合格类)为正例, 赋值为1, 多数类样本(合格类)赋值为0。初始数据集被划分为两部分, 一部分经数据重采样方法扩增, 用于生成训练集和验证集。训练集用于基模型PSO算法参数优化过程, 验证集则用于验证优化后基模型的性能; 另一部分作为测试集, 用于测试PSO-Stacking模型的预测能力, 评估模型的整体性能。最终, 训练集样本数量为3012件, 包含2978件合格与34件不合格; 验证集样本数量为751件, 包含742件合格与9件不合格; 测试集样本数量为126件, 包含114件合格与12件不合格。
本研究选取宏平均方法计算评价模型性能的指标[29]: 精准率、召回率、F1分数、AUC值。该方法首先计算每一类的指标值, 再求平均值, 适用于类别数量差异较大的不均衡数据集, 能够更好地反映小类别(不合格类)的分类性能。
宏平均方法计算得到的精确率指模型对正例的预测准确程度; 召回率表示实际正例被预测为正例的概率; F1分数是精确率和召回率的加权调和平均, 衡量二分类模型的精确度, 取值范围为0~1, 其值越趋近于1, 则说明模型精确度越高。计算公式见式(2)~(4)。
$P=\frac{\sum_{i=1}^{n} \frac{\mathrm{TP}_{i}}{\mathrm{TP}_{i}+\mathrm{FP}_{i}}}{n}$
$R=\frac{\sum_{i=1}^{n} \frac{\mathrm{TP}_{i}}{\mathrm{TP}_{i}+\mathrm{FN}_{i}}}{n}$
$F 1=\frac{\sum_{i=1}^{n} 2 \frac{R_{i} \times P_{i}}{R_{i}+P_{i}}}{n}$
式中: TP代表实际为合格, 预测也为合格的样本数; TN代表实际为不合格, 预测也为不合格的样本数; FP代表实际为合格, 预测为不合格的样本数; FN代表实际为不合格, 预测为合格的样本数。
受试者工作特征曲线(ROC曲线)以假阳性率(false positive rate, FPR)和真阳性率(true positive rate, TPR)为横轴、纵轴, 展示模型在所有可能阈值下的分类性能, 曲线距基线越远则说明模型性能越好。AUC值为ROC曲线下方的面积, 取值范围为0~1, 越趋近于1表示模型性能越佳。精确率-召回率曲线(precision-recall curve, P-R)反映了不同分类阈值下模型的精确率和召回率之间的关系, 适用于类别不均衡的情景, 平均精度(average precision, AP)是P-R下的面积, 反映模型的性能。
采用沙普利加性解释(shapley additive explanations, SHAP)方法解释PSO-Stacking模型的结果, 获得每个特征对模型预测结果的重要性[33]
运用WPS Office v13.19.0软件统计并分析蔬菜中倍硫磷的抽检数据, 在Python 3.11.4环境下建立基于蔬菜中倍硫磷抽检结果的PSO-Stacking预测模型。
2021—2023年共抽检上海市市售蔬菜样品3889件, 有55件样品中倍硫磷的残留量超过限量值(0.05 mg/kg), 总体不合格率为1.4% (55/3889)。按照蔬菜的品种分析, 豆类蔬菜中倍硫磷的不合格率最高(2.3%), 其次为茄果类蔬菜(0.2%), 瓜类蔬菜、叶菜类蔬菜、鳞茎类蔬菜、水生类蔬菜、根茎类和薯芋类蔬菜中倍硫磷均未超标, 详见表1
蔬菜中倍硫磷样品的抽样月份覆盖了全年12个月。其中, 2月抽样的蔬菜中倍硫磷的不合格率最高(12.6%), 其次为3月(7.7%)和11月(2.1%), 1月、6月、7月和12月蔬菜中倍硫磷均未超标, 详见表2
流通环节蔬菜中倍硫磷的不合格率为1.5% (50/3330), 其明显高于餐饮环节(0.9%)。抽样场所为网购的蔬菜中倍硫磷的不合格率最高(2.9%), 其次为小型餐馆(2.5%)、超市(1.9%)和食品商店(含食杂店)(1.9%), 商场、流通环节(其他)、大型餐馆和快餐店蔬菜中倍硫磷均未超标, 详见表3
以AUC值为目标函数的十折交叉验证结果如图2所示, 该图对比了模型与重采样方法的分类性能。图2A显示极端梯度提升和类别特征梯度提升树表现优异(AUC均>0.84), 而决策树表现较差(AUC为0.5814)。图2B表明ADASYN和SMOTE效果较好(AUC为0.8137和0.7815), ROS最低(AUC为0.6094)。选择AUC值大于0.8的机器学习模型, 包括XGBoost、CatBoost、RF、LGBM和GB, 结合ADASYN过采样技术构建Stacking预测模型。
通过PSO算法寻找得到5个基模型的最佳参数, 其中所有基模型共有的3个主要参数如表4所示。由表4可知: GB模型树的数量最多, 内存需求与计算量最大; RF模型树的数量最少, 模型复杂度较小。
随后, 分别构建最佳参数的基模型, 采用训练集单独训练基模型, 使用经验证集测试得到性能指标较优的基模型构建Stacking模型。为进一步验证模型的稳健性, 使用测试集数据测试所有模型性能, 计算相关的模型评价性能指标, 如表5所示。
表5可知, PSO-RF和PSO-GB的表现较差, 模型预测结果的精确率都低于0.5, 说明模型将更多的负例误判为正例; PSO-XGBoost和PSO-LGBM的分数接近, 后者表现略好, AUC值达到0.9; PSO-CatBoost模型的表现最佳, 其精确率、召回率和F1值均为0.83, AUC值为0.91。因此, 选择PSO-CatBoost模型作为Stacking模型的元模型, 其余为基模型, 构建PSO-Stacking模型。
构建的PSO-Stacking模型相关性能评价指标计算结果如表5所示, 表现较好。该模型的精确率为0.91, 召回率为0.83, F1值为0.87, AUC值为0.91。与单一的基模型相比, 其精确度有明显的提升, 提高了8.79%~52.75%; 召回率仅比PSO-XGBoost模型提高了9.64%; F1值和AUC值分别提高了4.60%~34.48%、1.10%~5.49%。提高了9.64%~111.63%; 召回率仅比PSO-XGBoost模型提高了10.67%; F1值和AUC值分别提高了4.82%~52.63%、0%~5.81%。
PSO-Stacking模型的混淆矩阵如图3A所示, 模型正确预测了10件不合格样本, 1件合格样本被误判为不合格样本, 精确率高达91%。P-R曲线如图3B所示, 该模型的AP值为0.77, 表明模型的性能较好; P-R曲线的拐点为(0.83,0.91), 在拐点前精确率随着召回率的升高而缓慢下降, 在拐点之后急剧下降, 该点即为模型寻找到精准率和召回率达到最佳的平衡点。
ROC曲线如图4所示, PSO-Stacking模型的性能略优于单一基模型, 拐点靠近(0,1)表明模型适用于蔬菜中倍硫磷的抽检结果预测。从整体看, 模型对抽检结果的预测准确度较高, 能有效区分正例样本(不合格样本), 评价指标在数量极少的正例样本上保持较高水平, 表明所建立的PSO-Stacking模型能够满足对蔬菜中倍硫磷不合格样本抽检结果预测的精度需求。
图5展示了基于SHAP值的样本聚类特征分布热力图, 反映了各样本在不同特征维度下对模型预测结果的贡献。热力图中的黑色曲线表示模型预测的f(x)值, 显示了样本序列的预测结果, 波动情况反映了不同样本在质量检测中的不合格情况。具体而言, 蔬菜品种的SHAP值分布偏向红色, 表明其对模型贡献最大; 检测机构和抽样场所的SHAP值次之, 部分样本在这两个特征上表现出负SHAP值, 提示检测流程或环境条件可能对结果产生消极影响; 任务类型的SHAP值几乎完全为负, 说明其贡献最小。整体热力图显示, 不合格样本与特定特征组合(如某种蔬菜品种在特定检测机构上)密切相关, 强调了特征间的关联性。此外, 结合样本的抽样月份, 热力图中的颜色变化显示不合格样本主要集中在2—4月和9—11月, 可能与蔬菜的生长季节等因素有关。
图6展示了单个样本预测为正例(不合格)的微观特征影响。随机从原始数据集中挑选4个抽检样本进行模型预测和分析。图6A为实际抽检结果为合格, 模型预测结果也为合格, 在数据集中按照标签对应的特征筛选后得到结果均为合格样本。图6B为实际不合格样本, 模型预测结果也为不合格。根据红色标签对应的特征筛选, 得到32个样本(包含8个不合格), 继续筛选蓝色标签对应的特征后, 剩余7个样本(包含1个不合格)。根据原始数据, 该样本在2月份通过网购从长宁区的流通环节抽取, 经检测机构24检测为不合格。图6B显示, 模型预测结果为不合格产生正向影响最大的特征是抽样月份(2月份), 产生负向影响的特征是蔬菜品种(茄果类)。因此, 食品安全决策者应重点关注来自2月份网购, 经检测机构24检测的蔬菜抽检样品。由图6C同样可知, 抽样地区(浦东新区)、抽样环节(流通环节)对预测结果为不合格产生负向影响。
本研究统计分析了2021—2023年3889件上海市市售蔬菜中倍硫磷的抽检数据, 发现豆类蔬菜中倍硫磷的不合格率最高(2.3%), 其次为茄果类蔬菜(0.2%), 总不合格率为1.4%。本研究建立的PSO-Stacking预测模型性能良好, 能够较准确的预测蔬菜抽检结果中不合格的样本。SHAP特征分布热力图有效揭示了蔬菜倍硫磷抽检数据的复杂性, 突出了影响模型预测结果的关键特征, 并为后续改进质量控制和检测流程提供了重要依据。
运用该模型, 食品监管部门可以输入与蔬菜抽检相关的特征信息输, 以预测蔬菜中倍硫磷的抽检结果。通过SHAP分析的热力图和微观样本特征影响图, 可以获得特征重要性排名以及各特征对结果的正向或负向影响。针对不合格样品, 监管部门能够迅速聚焦于对预测结果产生正向影响的特征变量, 便于食品安全决策者迅速做出反应, 从而调整监测重心, 以达到事前预警的目的。本研究构建的PSO-Stacking模型因其灵活性高, 适用于高维数据和不平衡数据, 通过将PSO和Stacking模型结合, 能够利用优化后的多个基模型提高模型的泛化能力, 为蔬菜监督抽检方案的制定提供参考, 提升蔬菜监督抽检的靶向性和问题食品处置的及时性。目前的研究仍有一些不足, 对于食品安全不均衡的抽检数据的预测仍然没有完美的解决方案。未来将着力于从算法层面深入挖掘, 以寻找不依赖数据重采样的高效建模方法, 实现更可靠的风险预警。
  • 2024年上海市市场监督管理局科技项目(2024-50)
  • 国家市场监督管理总局科技计划项目(2022MK034)
参考文献 引证文献
排序方式:
[1]
王芳, 李兆周, 李道敏, 等. 倍硫磷表面印迹聚合物的制备及表征[J]. 核农学报, 2016, 30(9): 1744-1753.
WANG F, LI ZZ, LI DM, et al. Preparation and characterization of fenthion surface molecularly imprinted polymers[J]. Journal of Nuclear Agricultural Sciences, 2016, 30(9): 1744-1753.
[2]
王子浩. 脂肪乳联合血液灌流对血浆中倍硫磷清除效果的评价[D]. 长春: 吉林大学, 2024.
WANG ZH. The evaluation of the clearance effect of fenthion in plasma by combining fat emulsion with hemoperfusion[D]. Changchun: Jilin University, 2024.
[3]
刘津, 甘兰萍, 孙通, 等. 激光诱导击穿光谱结合CARS变量选择方法定量检测倍硫磷含量[J]. 分析测试学报, 2017, 36(9): 1099-1103.
LIU J, GAN LP, SUN T, et al. Quantitative determination of fenthion using laser induced breakdown spectroscopy with CARS variable selection method[J]. Journal of Instrumental Analysis, 2017, 36(9): 1099-1103.
[4]
步岩刚, 钱宗耀. 倍硫磷在新疆哈密瓜中的残留检测及其在土壤中的降解行为研究[J]. 现代农药, 2024, 23(1): 70-74.
BU YG, QIAN ZY. Detection of fenthion residues in Hami melon and its degradation behavior in soils[J]. Modern Agrochemicals, 2024, 23(1): 70-74.
[5]
吴帅. 面向不平衡数据的食品安全风险预警技术研究[D]. 南宁: 广西大学, 2023.
WU S. Research on food safety risk prediction technology based on unbalanced data[D]. Nanning: Guangxi University, 2023.
[6]
WANG X, BOUZEMBRAK Y, OUDE-LANSINK AGJM, et al. Weighted Bayesian network for the classification of unbalanced food safety data: Case study of risk-based monitoring of heavy metals[J]. Risk Analysis, 2023, 43(12): 2549-2561.
[7]
LI T, ZUO E, CHEN C, et al. Gaussian distribution resampling via Chebyshev distance for food computing[J]. Applied Soft Computing, 2024, 150: 111103.
[8]
丁浩晗, 田嘉伟, 谢祯奇, 等. 机器学习和大数据在食品领域的应用[J]. 食品与发酵工业, 2024, 50(24): 353-361.
DING HH, TIAN JW, XIE ZQ, et al. Applications of machine learning and big data in food industry[J]. Food and Fermentation Industries, 2024, 50(24): 353-361.
[9]
DONG X, YU Z, CAO W, et al. A survey on ensemble learning[J]. Frontiers of Computer Science, 2020, 14(2): 241-258.
[10]
KEVIN M, CARLOS ND, RAZIUR R, et al. Investigation of model stacking for drug sensitivity prediction[J]. BMC Bioinformatics, 2018, 19(S3): 71.
[11]
SU L, LIAO H, YU Z, et al. Ensemble learning for question classification[Z]. 2009.
[12]
XU C, DING J, QIAO Y, et al. Tomato disease and pest diagnosis method based on the Stacking of prescription data[J]. Computers and Electronics in Agriculture, 2022, 197: 106997.
[13]
吕振哲. 基于重采样方法的不平衡数据分类研究[D]. 烟台: 烟台大学, 2024.
LV ZZ. Research on imbalanced data classification based on resampling methods[D]. Yantai: Yantai University, 2024.
[14]
ISLAM A, BELHAOUARI SB, REHMAN AU, et al. Knnor: An oversampling technique for imbalanced datasets[J]. Applied Soft Computing, 2022, 115: 108288.
[15]
BELHAOUARI SB, ISLAM A, KASSOUL K, et al. Oversampling techniques for imbalanced data in regression[J]. Expert Systems with Applications, 2024, 252: 124118.
[16]
周玉, 孙红玉, 房倩, 等. 不平衡数据集分类方法研究综述[J]. 计算机应用研究, 2022, 39(6): 1615-1621.
ZHOU Y, SUN HY, FANG Q, et al. Review of imbalanced data classification methods[J]. Application Research of Computer, 2022, 39(6): 1615-1621.
[17]
KHUSHI M, SHAUKAT K, ALAM TM, et al. A comparative performance analysis of data resampling methods on imbalance medical data[J]. IEEE Access, 2021, 9: 109960-109975.
[18]
LIU N, BOUZEMBRAK Y, VANDEN-BULK LM, et al. Automated food safety early warning system in the dairy supply chain using machine learning[J]. Food Control, 2022, 136: 108872.
[19]
梁玲玲. 食品安全风险预警方法研究及应用[D]. 北京: 北京化工大学, 2023.
LIANG LL. Research and application on food safety risk early warning method[D]. Beijing: Beijing University of Chemical Technology, 2023.
[20]
蔡照鹏, 徐林. 易腐食品安全预警云体系构建研究[J]. 河南城建学院学报, 2018, 27(6): 64-70.
CAI ZP, XU L. Early warning method of fresh chicken preservation period based on SVM[J]. Journal of Henan University of Urban Construction, 2018, 27(6): 64-70.
[21]
杨鸿雁, 田英杰. 机器学习在食品安全风险预警及抽检方案制订中的应用研究[J]. 管理评论, 2022, 34(11): 315-323.
YANG HY, TIAN YJ. Application research of machine learning in food safety risk early warning and sampling inspection program[J]. Management Review, 2022, 34(11): 315-323.
[22]
王小艺, 王姿懿, 赵峙尧, 等. 集成改进AHP与XGBoost算法的食品安全风险预测模型: 以大米为例[J]. 食品科学技术学报, 2022, 40(1): 150-158.
WANG XY, WANG ZY, ZHAO ZY, et al. A food safety risk forecast model integrated with improved AHP and XGBoost algorithm: A case study of rice[J]. Journal of Food Science and Technology, 2022, 40(1): 150-158.
[23]
高亚男, 王文倩, 王建新. 集成模糊层级划分的LightGBM食品安全风险预警模型: 以肉制品为例[J]. 食品科学, 2021, 42(1): 197-207.
GAO YN, WANG WQ, WANG JX. A food safety risk prewarning model using lightGBM integrated with fuzzy hierarchy partition: A case study for meat products[J]. Food Science, 2021, 42(1): 197-207.
[24]
卢清华. CatBoost算法在居民食品安全满意度调查中的应用—以上海市J区为例[D]. 烟台: 鲁东大学, 2023.
LU QH. Application of CatBoost algorithm in residents' food safety satisfaction survey—Take Shanghai J District as an example[D]. Yantai: Ludong University, 2023.
[25]
甄俊涛, 刘臣. 高维数据多标签分类的食品安全预警研究[J]. 计算机技术与发展, 2020, 30(9): 109-114.
ZHEN JT, LIU C. Research on food safety early warning of multilabel classification of high dimensional data[J]. Computer Technology and Development, 2020, 30(9): 109-114.
[26]
唐昊. 基于机器学习的典型食品安全风险预警研究[D]. 北京: 北京化工大学, 2023.
TANG H. Research on risk early warning of typical food safety based on machine learning[D]. Beijing: Beijing University of Chemical Technology, 2023.
[27]
MOAZEN H, MOLAEI S, FARZINVASH L, et al. PSO-ELPM: PSO with elite learning, enhanced parameter updating, and exponential mutation operator[J]. Information Sciences, 2023, 628: 70-91.
[28]
何浦. 基于粒子群算法的食品加工工艺参数优化模型构建[J]. 自动化与仪器仪表, 2022(10): 32-37.
HE P. Construction of optimization model of food processing parameters based on particles warm optimization[J]. Automation & Instrumentation, 2022(10): 32-37.
[29]
ZHOU ZW, TIAN ML, LIU BX, et al. Assessment of salmon sashimi processing conditions for Listeria monocytogenes cross-contamination and effectiveness of CLPSO-BP neural network model constructed for predicting microbial transfer[J]. LWT, 2024, 201: 116252.
[30]
邢巍. 基于人工神经网络和计算机视觉系统的烤肠多元安全品质预测模型构建[D]. 合肥: 合肥工业大学, 2023.
XING W. Construction of multivariate safety quality prediction model of roast sausage based on artificial neural network and computer vision system[D]. Hefei: Hefei University of Technology, 2023.
[31]
陈鹏飞. 基于SA-PSO-BP神经网络的牛肉微波干燥含水率预测研究[D]. 银川: 宁夏大学, 2023.
CHEN PF. Research on prediction of moisture content of beef microwave drying based on SA-PSO-BP neural network[D]. Yinchuan: Ningxia University, 2023.
[32]
GENG Z, WANG X, JIANG Y, et al. Novel IAPSO-LSTM neural network for risk analysis and early warning of food safety[J]. Expert Systems with Applications, 2023, 230: 120747.
[33]
宁霄, 贺鑫鑫, 金绍明, 等. 食品快速检测方法的评价及其贝叶斯推断[J]. 食品安全质量检测学报, 2021, 12(3): 1015-1020.
NING X, HE XX, JIN SM, et al. Evaluation of food rapid detection methods and their Bayesian inference[J]. Journal of Food Safety and Quality, 2021, 12(3): 1015-1020.
2025年第16卷第5期
PDF下载
275
99
引用本文
BibTeX
文章信息
doi: 10.19812/j.cnki.jfsq11-5956/ts.20241008004
  • 接收时间:2024-10-08
  • 首发时间:2025-07-19
  • 出版时间:2025-03-15
补充材料
相关文章
文章信息
作者
出版历史
  • 收稿日期:2024-10-08
基金
2024年上海市市场监督管理局科技项目(2024-50)
国家市场监督管理总局科技计划项目(2022MK034)
作者信息
    上海市市场监督管理局信息应用研究中心, 上海 200030

通讯作者:

* 彭少杰(1968—), 男, 硕士, 主任医师, 主要研究方向为食品安全风险监测与评估。E-mail:
参考文献
分享链接
https://castjournals.cast.org.cn/joweb/spaq/CN/10.19812/j.cnki.jfsq11-5956/ts.20241008004
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

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

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