Towards Text-Based Recommendations

Damien Poirier (1, 2) , Isabelle Tellier (2) , Françoise Fessant (1) , Julien Schluth (3)

1 Orange Labs [Lannion]
2 LIFO - Laboratoire d'Informatique Fondamentale d'Orléans
3 GFI Informatique

Recommender systems have become, like search engines, a tool that cannot be ignored by a website with a large selection of products, music, news or simply webpages. The performance of this kind of systems depends on a large amount of information. Meanwhile, the amount of information available in the Web is continuously growing. In this paper, we propose to provide recommendation from unstructured textual data. The method has two steps. First, subjective texts are labelled according to their expressed opinion. Second, the results are used to provide recommendations thanks to a collaborative filtering technique. We describe the complete processing chain and evaluate it.

Domains: Information Retrieval [cs.IR]

Damien Poirier, Isabelle Tellier, Françoise Fessant, Julien Schluth. Towards Text-Based Recommendations. RIAO 2010: 9th international conference on Adaptivity, Personalization and Fusion of Heterogeneous Information, Apr 2010, PARIS, France. pp.0-0. ⟨hal-00470172⟩