期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：8
DOI：10.15680/IJIRCCE.2015. 0308023
出版社：S&S Publications
摘要：Although personalized search has been discovered for many years and many personalization strategieshave been invented, it is still undecided whether personalization is consistently effective on different queries fordifferent users, and under dissimilar search contexts. According to these user comments, rating can be done about anyproduct launched in market, but it quite difficult. To assuage this problem, we propose a logistic regression algorithmwhich will gather round all the comments fetched by system witness and able to compute positive-negative probability,with the help of this probability we can rate movie, industrial company and education societies. This proposed systemdesigned on the basis of comments left by user earlier. An actual movie, industrial company, education societies reviewdata set achieved from a huge data portal, our system help to improve significance comparison performance. Theproposed algorithm can accurately predict a user‟s preferences in their interest area. Using online user survey we canget true user interest preference about multimedia as well as web mining content
关键词：Logistic Regression. Data mining; personalized search contents; user comments; Logistic regression;recommender system; Sentiment analysis
