A Cognitive Agent Framework in Information Retrieval : Using User Beliefs to Customize Results

Dima El Zein (1, 2, 3) , Célia da Costa Pereira (1, 2, 3)

1 I3S - Laboratoire d'Informatique, Signaux, et Systèmes de Sophia Antipolis
2 Laboratoire I3S - SPARKS - Scalable and Pervasive softwARe and Knowledge Systems
3 UniCA - Université Côte d'Azur

This paper presents a framework for a cognitive agent in information retrieval that customizes the list of returned documents based on what the agent believes about the user knowledge. Throughout the interactions between the agent and the user, the agent builds its beliefs by extracting the content of the documents examined by the user. The agent's belief base consists of "simple beliefs" represented by the document's keywords as well as "contextual rules" that allow the agent to derive new beliefs about the user knowledge. The agent is therefore able to compare its own beliefs about the user knowledge with the knowledge conveyed by a given document, and thus understand if the document really contains useful information for the user or not. Finally, in case of inconsistency, the agent revises its belief base to restore consistency.

Mots clés: Cognitive Agent; Information Retrieval; Knowledge Extraction; Belief Revision

Domaines: Informatique [cs]

Dima El Zein, Célia da Costa Pereira. A Cognitive Agent Framework in Information Retrieval : Using User Beliefs to Customize Results. The 23rd International Conference on Principles and Practice of Multi-Agent Systems, Nov 2020, Nagoya, Japan. ⟨10.1007/978-3-030-69322-0_21⟩. ⟨hal-03029127⟩