Personalized Ranking of Search Results with Implicitly Learned User Interest Hierarchies

Web search engines are usually designed to serve all users, without considering the interests of individual users. Personalized web search incorporates an individual user's interests when deciding relevant results to return. We propose to learn a user profile, called a user interest hierarchy (UIH), from web pages that are of interest to the user. The user’s interest in web pages will be determined implicitly, without directly asking the user. Using the implicitly learned UIH, we study methods that (re)rank the results from a search engine. Experimental results indicate that our personalized ranking methods, when used with a popular search engine, can yield more relevant web pages for individual users.