Information retrieval systems such as web search engines are
critical for overcoming information overload. A major deficiency of
existing retrieval systems is that they generally lack implicit user
modeling and thus not adaptive to individual users, resulting in
inherently non-optimal retrieval performance. The basic idea of this work
is to make use of the implicit relevance feedback present in the user's
interactions with search results in order to improve the web search
ranking and automatically augment queries that better describe the users
need. This is achieved using a client-side web search agent that resides
locally on the users machine. The tendency towards a client-side approach
provides privacy and ensures that all the users interactions are
recorded.