Post-Retrieval Clustering Using Third-Order Similarity Measures

José G. Moreno (1) , Gaël Dias (1) , Guillaume Cleuziou (2)

1 Equipe Hultech - Laboratoire GREYC - UMR6072
2 LIFO - Laboratoire d'Informatique Fondamentale d'Orléans

Post-retrieval clustering is the task of clustering Web search results. Within this context, we propose a new methodology that adapts the classical K-means algorithm to a third-order similarity measure initially developed for NLP tasks. Results obtained with the definition of a new stopping criterion over the ODP-239 and the MORESQUE golden standard datasets evidence that our proposal outperforms all reported text-based approaches.

Domains: Information Retrieval [cs.IR]; Machine Learning [cs.LG]

José G. Moreno, Gaël Dias, Guillaume Cleuziou. Post-Retrieval Clustering Using Third-Order Similarity Measures. Annual Meeting of the Association for Computational Linguistics (ACL 2013), Aug 2013, Sofia, Bulgaria. pp.153-158. ⟨hal-00931263⟩