Web Search and Recommendation-based on User Interest for PWS

  Arun R. Desai,  Mr. Pankaj Chandre

Numerous personalization approaches have been investigated but it is still unclear whether personalization is reliably effective on dissimilar queries for different users, and under different search contexts. Personalized web search (PWS) has established its effectiveness in increasing the quality of several search services on the web. This paper proposes a personalized web search (PWS) framework known as User customizable Privacy-preserving Search (UPS) that can adaptively specify profiles by queries while regarding user quantified privacy requirements. The system goals at striking a balance among two predictive metrics that estimate the utility of personalization as well as the privacy risk of exposing the generalized profile. The greedy algorithm namely GreedyIL is presented for runtime generalization. Additionally, paper provides an online web age prediction mechanism for deciding whether personalizing a query is helpful. Additionally, this paper proposes a Personalized Web page Recommendation model (PWR) through collaborative filtering and a topic-aware Markov model. Topic-aware Markov model is used to widely applied to learn users’ navigation behaviors for predicting the next step while surfing the Web.

Keywords- Privacy preservation, personalized web search, recommendation, profile privacy risk, user profile

Unique Identification Number - IJEDR1602341
Page Number(s) - 1939-1945
Pubished in - Volume 4 | Issue 2 | July 2016
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Arun R. Desai,  Mr. Pankaj Chandre,   "Web Search and Recommendation-based on User Interest for PWS", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.4, Issue 2, pp.1939-1945, July 2016, Available at :http://www.ijedr.org/papers/IJEDR1602341.pdf