资源描述:
《基于云模型的协同过滤推荐算法》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库。
1、ISSN1000-9825,CODENRUXUEWE-mail:jos@iscas.ac.cnJournalofSoftware,Vol.18,No.10,October2007,pp.2403−2411http://www.jos.org.cnDOI:10.1360/jos182403Tel/Fax:+86-10-62562563©2007byJournalofSoftware.Allrightsreserved.∗基于云模型的协同过滤推荐算法1,4+2312张光卫,李德毅,李鹏,康建初,陈桂生1(北京航空航天大学软件开发环境国家重点实验室,北京1
2、00083)2(中国电子工程系统研究所,北京100840)3(哈尔滨工业大学深圳研究生院信息安全中心,广东深圳518055)4(山东建筑大学计算机科学与技术学院,山东济南250101)ACollaborativeFilteringRecommendationAlgorithmBasedonCloudModel1,4+2312ZHANGGuang-Wei,LIDe-Yi,LIPeng,KANGJian-Chu,CHENGui-Sheng1(StateKeyLaboratoryofSoftwareDevelopmentEnvironment,BeiHan
3、gUniversity,Beijing100083,China)2(InstituteofElectronicSystemEngineering,Beijing100840,China)3(InformationSecurityCenter,ShenzhenGraduateSchool,HarbinInstituteofTechnology,Shenzhen518055,China)4(SchoolofComputerScienceandTechnology,ShandongJianzhuUniversity,Ji’nan250101,China)+
4、Correspondingauthor:Phn:+86-10-82316732,E-mail:ezhang@263.net,http://nlsde.buaa.edu.cnZhangGW,LiDY,LiP,KangJC,ChenGS.Acollaborativefilteringrecommendationalgorithmbasedoncloudmodel.JournalofSoftware,2007,18(10):2403−2411.http://www.jos.org.cn/1000-9825/18/2403.htmAbstract:Recom
5、mendationsystemisoneofthemostimportanttechnologiesappliedine-commerce.Similaritymeasuringmethodisfundamentaltocollaborativefilteringalgorithm,andtraditionalmethodsareinefficientespeciallywhentheuserratingdataareextremelysparse.BasedontheoutstandingcharacteristicsofCloudModelont
6、heprocessoftransformingaqualitativeconcepttoasetofquantitativenumericalvalues,anovelsimilaritymeasuringmethod,namelythelikenesscomparingmethodbasedoncloudmodel(LICM)isproposedinthispaper.LICMcomparesthesimilarityoftwousersonknowledgelevel,whichcanovercomethedrawbackofattributes
7、’strictlymatching.Thisworkanalysistraditionalmethodsthroughlyandputsforwardanovelcollaborativefilteringalgorithm,whichisbasedontheLICMmethod.ExperimentsontypicaldatasetshowtheexcellentperformanceofthepresentcollaborativefilteringalgorithmbasedonLICM,evenwithextremelysparsityofd
8、ata.Keywords:cloudmodel;collaborativefiltering;similar