Finding People and Documents, Using Web 2.0 Data

Given a user’s free-text query, search engines return a ranked list of documents that are likely to be helpful to the user. In this research, we propose a simple yet highly effective technique for also providing a ranked list of related people to every search. The list of people related to the query is calculated at search time using an enhanced faceted search engine, based on person-document relationships mined from several Web 2.0 applications (such as blogs and social bookmarks) in the intranet of a large enterprise. Our hypothesis is that the related people we retrieve for a query are people who have special interest in the query’s topic, and thus may be useful to the person making this query. We conducted a large user study with over 600 people to confirm this hypothesis.