A Contextual Information Retrieval Model based on Influence Diagrams

Nesrine Zemirli (1) , Lynda Tamine (1) , Mohand Boughanem (1)

1 IRIT-SIG - Systèmes d’Informations Généralisées

A key challenge in information retrieval is the use of contex- tual evidence within the ad-hoc retrieval. Our contribution is particularly based on the belief that contextual retrieval is a decision-making prob- lem. For this reason we propose to apply influence diagrams witch are an extension of Bayesian networks to such problems, in order to solve the hard problem of user based relevance estimation. The basic underlying idea is to substitute to the traditional relevance function which measures the degree of matching document-query, a function indexed by the user. In our approach, the user profile is represented by his long term interests. In order to validate our model, we propose furthermore a novel evaluation protocol suitable for the contextual retrieval task. The test collection is an expansion of the standard TREC test data, obtained using a learning scenario of the user's interests. The experimental results show that our model is promising.

Domaines: Recherche d'information [cs.IR]

Nesrine Zemirli, Lynda Tamine, Mohand Boughanem. A Contextual Information Retrieval Model based on Influence Diagrams. 2nd International Workshop on Context Based Information Retrieval (2007), Aug 2007, Roskilde, Denmark. ⟨hal-00359546⟩