PLSA-Based Personalized Information Retrieval with Network Regularization

WiththerapiddevelopmentoftheInternet,theinformationretrievalmodelbasedonthekeywords matchingalgorithmhasnotmettherequirementsofusers,becausepeoplewithvariousqueryhistory alwayshavedifferentretrievalintentions.Userqueryhistoryoftenimpliestheirinterests.Therefore,it isofgreatimportancetoenhancetherecallratioandtheprecisionratiobyapplyingqueryhistoryinto thejudgmentofretrievalintentions.Forthissake,thisarticledoesresearchonuserqueryhistoryand proposesamethodtoconstructuserinterestmodelutilizingqueryhistory.Coordinately,theauthors designamodelcalledPLSA-basedPersonalizedInformationRetrievalwithNetworkRegularization. Finally,themodelisappliedintoacademicinformationretrievalandtheauthorscompareitwith BaiduScholarand thepersonalized informationretrievalmodelbasedon theprobabilistic latent semanticanalysistopicmodel.Theexperimentresultsprovethatthismodelcaneffectivelyextract topicsandretrievesbackresultsmoresatisfiedforusers’requirements.Also,thismodelimproves theeffectofretrievalresultsapparently.Inaddition,theretrievalmodelcanbeutilizednotonlyin theacademicinformationretrieval,butalsointhepersonalizedinformationretrievalonmicroblog search,associaterecommendation,etc. KeywoRdS Network Regulation, Personalized Information Retrieval, PLSA, Query History, Query Intention