The personalization process can either be focused on individuals and their interaction with documents, or on the identification of shared patterns of behavior and the segmentation of the user population into groups of common interests. One issue in the personalization and the filtering processes is the selection of an appropriate model for the efficient representation and manipulation of user profiles and documents. It should be capable of facilitating the determination of relevant documents in terms of similarity between users and documents. The aim of this paper is to introduce three models for representing documents and profiles in the search process, and to examine their computational processes. The volume of document databases, the large number of users and their different interests creates the need for precise and efficient filtering techniques. This paper investigates different information retrieval models, which can be used to determine the similarity between documents and user profiles.
