Abstract
Based on user interest and information requirement Personalized Web Search (PWS) delivers different search results for disguised users. Personalized web search have disguise characteristics while compared with common web search, as which deliver same set of search result for the same keyword search, by different kind of user have different needs. Really, these diligences have become one of the main hurdles for locating personalized search and how to do privacy-preserving personalization is a extensive challenge. Hence to overcome these difficulties privacy protection in Personalized Web Search provides a model hierarchical user profile, which have been built based on user preferences. Propose a PWS framework User Customizable Online Privacy-preserving Search with K-anonymity (UCOPSK) which generalizes profile as per the user specified privacy requirements in online and offline search. In this proposed work the profiles are constructed for each static and dynamic user in the websites. K-anonymity is applied to each user profile to manifest of sensitive information of user in privacy preservation, which can significantly prevent the sensational information leakage under attacks, and it is commonly used in discrete fields now a days. This paper describes the various approaches and techniques of preserving user data applied on personalized web search to build up a new algorithm & method to improve performance, utility and security of existing data and help to create the new predictions on the data. This paper describes the comparative study of clustering techniques used to improve privacy preservation on personalized web search.
View more >>