Query expansion in information retrieval : What can we learn from a deep analysis of queries ?

Liana Ermakova (1) , Josiane Mothe (1, 2) , Irina Ovchinnikova (3)

1 IRIT-SIG - Systèmes d’Informations Généralisées
2 UT2J - Université Toulouse - Jean Jaurès
3 Perm State University

Information retrieval aims at retrieving relevant documents answering a user's need expressed through a query. Users' queries are generally less than 3 words which make challenging to answer correctly. Automatic query expansion (QE) improves the precision in average even if it can decrease the results for some queries. In this paper, we propose a new automatic QE method that estimates the importance of expansion candidate terms by the strength of their relation to the query terms. The method combines local analysis and global analysis of texts. We evaluate the method using international benchmark collections and measures. We found comparable results in average compared to the Bo2 method. However, we show that a deep analysis of initial and expanded queries brings interesting insights that could help for future research in the domain.

Mots clés: Information retrieval; Analysis of query terms; Query expansion; Global analysis; Relevance feedback; Co-occurrence

Domaines: Théorie de l'information [cs.IT]; Recherche d'information [cs.IR]

Liana Ermakova, Josiane Mothe, Irina Ovchinnikova. Query expansion in information retrieval : What can we learn from a deep analysis of queries ?. International Conference on Computational Linguistics - Dialogue 2014, Jun 2014, Moscow, Russia. pp. 162-172. ⟨hal-01142602⟩