Recommending Personalized Search Terms for Assisting Exploratory Website Search

When visiting a website, the main task is to search for relevant information available on the site. The search is typically conducted using search terms. Choosing the right search terms is crucial to a successful search, but is often difficult, especially for exploratory searches. In this paper, we propose a recommender system called mySearchCLUE that recommends personalized search terms within a website search engine via collaborative filtering of similar visitors. The proposed recommender system helps website visitors explore and find better personalized search terms compared with no recommendations, and thus can be used to aid the personalized lookup and/or exploratory search process.