Multiview Semi-Supervised Learning for Ranking Multilingual Documents

Nicolas Usunier (1) , Massih-Reza Amini (1) , Cyril Goutte

1 MALIRE - Machine Learning and Information Retrieval

We address the problem of learning to rank documents in a multilingual context, when reference ranking information is only partially available. We propose a multiview learning approach to this semi-supervised ranking task, where the translation of a document in a given language is considered as a view of the document. Although both multiview and semi-supervised learning of classifiers have been studied extensively in recent years, their applicatin to the problem of ranking has received much less attention. We describe a semi-supervised multi-veiw ranking algorithm that exploits a global agreement between view-specific ranking functions on a set of unlabeled observations. We show that our proposed algorithm achieves significant improvements over both semi-supervised multiview classification and semi-supervised single-view rankers on a large multilingual collection of Reuters news covering 5 languages. Our experiments also suugest that our approach is most effective when few labeled documents are available and the classes are imbalanced.

Domaines: Informatique [cs]

Nicolas Usunier, Massih-Reza Amini, Cyril Goutte. Multiview Semi-Supervised Learning for Ranking Multilingual Documents. European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2011, Athens, Greece. pp.443-458, ⟨10.1007/978-3-642-23808-6_29⟩. ⟨hal-01286156⟩