Leyla Zhuhadar, Olfa Nasraoui, Robert Wyatt
Several semantic based user profile approaches have been in- troduced in the literature to learn the users' interests for personalized search. However, many of them are ill-suited to cope with a domain of information that evolves and user interests that may change over time. In this paper, we propose a novel dual representation of a user's semantic profile to deal with this problem: (1) a lower-level semantic representation, consisting of an accumulated gathering of user activities over a long period of time, that uses a standard machine learning algorithm to detect user convergence, (2) a higher-level semantic representation that detects shifts in the user activitiesonce this shift is detected, the higher-level semantic representation automatically updates the user profiles and reinitialize the system. Our experimental results demonstrate the feasibility of this approach.