返回
a hybrid of sequential rules and collaborative filtering for product recommendation.pdf

a hybrid of sequential rules and collaborative filtering for product recommendation.pdf

ID：33942559

大小：472.88 KB

页数：15页

时间：2019-02-28

a hybrid of sequential rules and collaborative filtering for product recommendation.pdf_第1页
a hybrid of sequential rules and collaborative filtering for product recommendation.pdf_第2页
a hybrid of sequential rules and collaborative filtering for product recommendation.pdf_第3页
a hybrid of sequential rules and collaborative filtering for product recommendation.pdf_第4页
a hybrid of sequential rules and collaborative filtering for product recommendation.pdf_第5页
资源描述：

《a hybrid of sequential rules and collaborative filtering for product recommendation.pdf》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、InformationSciences179(2009)3505–3519ContentslistsavailableatScienceDirectInformationSciencesjournalhomepage:www.elsevier.com/locate/insAhybridofsequentialrulesandcollaborativefilteringforproductrecommendationDuen-RenLiu*,Chin-HuiLai,Wang-JungLeeInstituteofInformationManagement,National

2、ChiaoTungUniversity,1001TaHseuhRoad,Hsinchu30050,TaiwanarticleinfoabstractArticlehistory:Customers’purchasebehaviormayvaryovertime.Traditionalcollaborativefiltering(CF)Received27June2007methodsmakerecommendationstoatargetcustomerbasedonthepurchasebehaviorReceivedinrevisedform13May2009of

3、customerswhosepreferencesaresimilartothoseofthetargetcustomer;however,Accepted5June2009themethodsdonotconsiderhowthecustomers’purchasebehaviormayvaryovertime.Incontrast,thesequentialrule-basedrecommendationmethodanalyzescustomers’pur-chasebehaviorovertimetoextractsequentialrulesinthefo

4、rm:purchasebehaviorinpre-viousperiods)purchasebehaviorinthecurrentperiod.Ifatargetcustomer’spurchaseKeywords:behaviorhistoryissimilartotheconditionalpartoftherule,thenhis/herpurchasebehav-CollaborativefilteringCustomersegmentationiorinthecurrentperiodisdeemedtobetheconsequentpartoftheru

5、le.AlthoughtheProductrecommendationsequentialrulemethodconsidersthesequenceofcustomers’purchasebehaviorovertime,Sequentialruleitdoesnotutilizethetargetcustomer’spurchasedataforthecurrentperiod.Toresolvetheaboveproblems,thisworkproposesanovelhybridrecommendationmethodthatcom-binestheseg

6、mentation-basedsequentialrulemethodwiththesegmentation-basedKNN-CFmethod.Theproposedmethodusescustomers’RFM(Recency,Frequency,andMonetary)valuestoclustercustomersintogroupswithsimilarRFMvalues.Foreachgroupofcustomers,sequentialrulesareextractedfromthepurchasesequencesofthatgrouptomaker

7、ecommendations.Meanwhile,thesegmentation-basedKNN-CFmethodprovidesrecommendationsbasedonthetargetcustomer’spurchasedataforthecurrentperiod.Then,theresultsofthetwomethodsarecombinedtomakefinalrecommendations.Exper-imentresultsshowthatthehybridmethodoutperformstraditionalCFmethods.Ó2009

当前文档最多预览五页，下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页，下载文档查看全文
温馨提示：
1. 部分包含数学公式或PPT动画的文件，查看预览时可能会显示错乱或异常，文件下载后无此问题，请放心下载。
2. 本文档由用户上传，版权归属用户，天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容，确认文档内容符合您的需求后进行下载，若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误，付费完成后未能成功下载的用户请联系客服处理。