J.-P. Sansonnet

Personalized Access to Information by Query Reformulation Based on the State of the Current Task and User Profile

Access to relevant information adapted to the needs and context of the user is a main challenge in the Internet world. Recent studies have dynamically enhanced the user query with the related users preferences, which are stored in users profile for providing personalized results; however, not all user preferences are relevant in all situations. As far as we know, these studies don’t take into account the state of the current activity during the information retrieval process. This paper presents a method to reformulate user queries depending on his profile together with the state of the task that he is undertaking when the information retrieval process takes places. Here suppose, that the queries we consider are related to some current task at hand. The UML state diagram is chosen for modeling the current task in order to detect the transitions at time intervals with the task state changes. In this paper, a new concept of SRQ (State Reformulated Queries) is introduced and used to reformulate queries. The result obtained by SRQ is more relevant than that obtained by the initial query for the current task at hand.