Anonymous personalization in collaborative web search

  • Authors:
  • Barry Smyth;Evelyn Balfe

  • Affiliations:
  • Smart Media Institute, University College Dublin, Dublin 4, Ireland;School of Computer Science and Informatics, College of Engineering, Mathematical and Physical Sciences, University College Dublin (UCD), Dublin 4, Ireland

  • Venue:
  • Information Retrieval
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present an innovative approach to Web search, called collaborative search, that seeks to cope with the type of vague queries that are commonplace in Web search. We do this by leveraging the search behaviour of previous searchers to personalize future result-lists according to the implied preferences of a community of like-minded individuals. This technique is implemented in the I-SPY meta-search engine and we present the results of a live-user trial which indicates that I-SPY can offer improved search performance when compared to a benchmark search engine, across a variety of performance metrics. In addition, I-SPY achieves its level of personalization while preserving the anonymity of individual users, and we argue that this offers unique privacy benefits compared to alternative approaches to personalization.