Toward a Context-Aware Multilingual Personalized Search

In recent years, personalized search has widely been used in Information Retrieval Systems (IRS) to provide the end user with more sophisticated and accurate search results. A basic element that plays an important role in personalized search is the user context which contains several aspects such as the user preferences, navigation history, habits, etc. A user may express his information needs in various languages. This requires the IRS to be able to consider all the contextual information provided in these languages. In this work, we present M-CAIRS, a Multilingual Context-aware Information Retrieval System that takes into account multilingual user contexts to better model the user search interests. Experimental results show a strong correlation between the user's relevance judgment and the automatic results obtained by our system, which proves the consistency and adequacy of our proposal.