Are Semantically Coherent Topic Models Useful for Ad Hoc Information Retrieval?

Romain Deveaud (1) , Eric Sanjuan (1) , Patrice Bellot (2)

1 LIA - Laboratoire Informatique d'Avignon
2 LSIS - Laboratoire des Sciences de l'Information et des Systèmes

The current topic modeling approaches for Information Retrieval do not allow to explicitly model query-oriented latent topics. More, the semantic coherence of the topics has never been considered in this field. We propose a model-based feedback approach that learns Latent Dirichlet Allocation topic models on the top-ranked pseudo-relevant feedback, and we measure the semantic coherence of those topics. We perform a first experimental evaluation using two major TREC test collections. Results show that retrieval performances tend to be better when using topics with higher semantic coherence.

Domaines: Informatique [cs]

Romain Deveaud, Eric Sanjuan, Patrice Bellot. Are Semantically Coherent Topic Models Useful for Ad Hoc Information Retrieval?. ACL 51st ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS, Aug 2013, Sofia, Bulgaria. ⟨hal-01314959⟩