Processing Natural Language Queries to Disambiguate Named Entities and Extract Users' Goals : Application to e-Tourism

This paper presents a study which is part of a broader project. This latter aims at providing mobile users with context-aware personalised services. The E-tourism Project deals with a variety of queries submitted by a tourist, such as booking a hotel room, getting the weather conditions for the next day, or booking tickets in a museum in the neighbourhood and worth to visit. This paper focuses on the query management and processing. The module described analyses and structures the query by splitting it, identifying the named entities, solving ambiguities... To process the query, the system uses various external knowledge bases and Natural Language Processing tools to understand the named entities and proper context of the query using disambiguation techniques. MOTS-CLÉS : e-Tourism, Traitement de Requêtes, Services Personnalisés Dépendants du Contexte