A self-adaptive classification for the dissociating privacy agent

This paper describes an extension of the Dissociating Privacy Agent (DisPA), which is a Privacy Enhancement Technology (PET) for web search. It is implemented as an add-on for Firefox that acts like a proxy between the user and the search engine. A fundamental part of DisPA is the classification of queries into a set of categories. Particularly, the taxonomy of the Open Directory Project (ODP) is used for this purpose. In this paper we briefly recall the internal operations of the agent, discuss the drawbacks of the current model and propose an improvement to overcome them.