Evolving Disjunctive and Conjunctive Topical Queries based on Multi-objective Optimization Criteria

CECCHINI ROCIO LUJÁN; LORENZETTI CARLOS MARTÍN; MAGUITMAN ANA GABRIELA

INTELIGENCIA ARTIFICIAL. IBERO-AMERICAN JOURNAL OF ARTIFICIAL INTELLIGENCE

AEPIA

Lugar: Madrid; Año: 2009 vol. 13 p. 14 - 26

1137-3601

In this work we propose techniques based on single- and multi-objective evolutionary algorithms to automatically evolve a population of topical queries. The developed techniques can be applied in the implementation of a topical search system.  We report on the results of different strategies that attempt to evolve conjunctive and disjunctive queries. Our analysis reveals the limitations of the single-objective approach and highlights the advantages of applying multi-objective evolutionary algorithms for the problem at hand. In addition,  we observe that disjunctive queries have the potential to achieve better retrieval performance than conjunctive queries. Finally, we show that the multi-objective evolutionary approach results in better performance than a baseline and other state-of-the-art techniques for query refinement.