Supporting Privacy Protection in Personalized Web Search

Abstract

The potency of Personalized web search (PWS) inenhancing the quality of diverse search services on the Internetis authenticated. Nevertheless, user’s disinclination to unfold their private information in the course of their search has created a vitalstop for the proliferation of PWS. We aspire to propose a PWS framework called UPS. while valuing user specified privacy requirements, with the help of queries, this can adaptively generalize profiles.This technique aims at maintaining equilibrium between two predictive metrics that gauges the utility of personalization and the privacy risk of uncovering the generalized profile. GreedyDP and GreedyIL are the two greedy algorithms for runtime generalization are proposed here. Additionally, weimpart an online prediction mechanism for deciding whether personalizing a query is serviceable. Extensive experiments demonstrate the effectiveness of our framework. The experimental results also reveal that GreedyIL outstandingly surpasses GreedyDP in terms of efficiency.