Usage Data in Web Search: Benefits and Limitations

Abstract

Web Search, which takes its root in the mature field of information retrieval, evolved tremendously over the last 20 years. The field encountered its first revolution when it started to deal with huge amounts of Web pages. Then, a major step was accomplished when engines started to consider the structure of the Web graph and link analysis became a differentiator in both crawling and ranking. Finally, a more discrete, but not less critical step, was made when search engines started to monitor and mine the numerous (mostly implicit) signals provided by users while interacting with the search engine. We focus here on this third “revolution” of large scale usage data. We detail the different shapes it takes, illustrating its benefits through a review of some winning search features that could not have been possible without it. We also discuss its limitations and how in some cases it even conflicts with some natural users’ aspirations such as personalization and privacy. We conclude by discussing how some of these conflicts can be circumvented by using adequate aggregation principles to create “ad hoc” crowds.