Научны(е)й руководител(и)ь : Patrick GALLINARI

Со-руководитель : USUNIER Nicolas

Learning-to-Rank consistent surrogates for Information Retrieval tasks

In this era of technology, managing, controlling and retrieving informations sources has become a key part of our daily lives, and at the same time it presents a crucial challenge to researchers. In this thesis we tackle the problem of searching for items from a whole collection of objects, this is usually called Information Retrieval. We address in a Learning-to-Rank fashion where the goal is to learn a scoring function while minimizing a risk which reflects the quality of the ranked list. In practice, this risk cannot be directly optimized and the main goal is to design related surrogate losses thanks to the consistency property. Starting from this property, we show a way to derive two consistent surrogate loss functions with respect to some traditional Information Retrieval metrics. The resultant experiments prove our methodology.
All these theoretical considerations cannot be directly applied due to the fact that learning algorithms are sensitives to the data quality. We focus our attention on the preprocessing that needs to be done on the data to make Learning-to-Rank algorithms efficient in a two case study: XML retrieval and Text-Image Retrieval problems. For both, Learning-to-Rank algorithms are dependent on the quality of the supervision, the sampling of the training examples and the extracted features describing them. Finally, we conduct a series of experiments on these two problems to outperform traditional state-of-the-art Information Retrieval methods.

Публикации 2008-2015

• 2015
  • G. Silvestre, D. Buffoni, K. Pires, S. Monnet, P. Sens : “Boosting Streaming Video Delivery with WiseReplica”, Transactions on Large-Scale Data- and Knowledge-Centered Systems, vol. XX, pp. 34-58, (Springer Berlin / Heidelberg) (2015)
• 2013
  • G. Silvestre, S. Monnet, D. Buffoni, P. Sens : “Predicting Popularity and Adapting Replication of Internet Videos for High-Quality Delivery”, Parallel and Distributed Systems (ICPADS), 2013 International Conference on, Seoul, Korea, Republic of, pp. 412-419, (IEEE) (2013)
• 2012
  • D. Buffoni : “Learning-to-Rank consistent surrogates for Information Retrieval tasks”, диссертация, Защита диссертаций 04.10.2012, Научны(е)й руководител(и)ь Gallinari, Patrick, Со-руководитель : Usunier, Nicolas (2012)
  • D. Buffoni, S. Tollari, P. Gallinari : “A Learning to Rank framework applied to text-image retrieval”, Multimedia Tools and Applications, vol. 60 (1), pp. 161-180, (Springer Verlag) (2012)
• 2011
  • D. Buffoni, S. Tollari, P. Gallinari : “The importance of the depth for text-image selection strategy in Learning to Rank”, European Conference on Information Retrieval (ECIR 2011), vol. 6611, Lecture Notes in Computer Science, Dublin, Ireland, pp. 743-746, (Springer) (2011)
  • D. Buffoni, C. Calauzènes, P. Gallinari, N. Usunier : “Learning Scoring Functions with Order-Preserving Losses and Standardized Supervision”, The 28^th International Conference on Machine Learning (ICML 2011), Bellevue, WA, United States, pp. 825-832 (2011)
  • T. Truong, D. Buffoni, N. Usunier, M.‑R. Amini, P. Gallinari : “Modèles d’Ordonnancement pour le Résumé Automatique et la Recherche d’Information”, chapitre de Modèles Statistiques pour l'Accès à l'Information Textuelle, pp. 19-42, (Hermès-Lavoisier), (ISBN: 2746224976) (2011)
• 2010
  • D. Buffoni, N. Usunier, P. Gallinari : “Apprentissage de fonctions d’ordonnancement par classification de paires ordonnées et pondérées (OWPC)”, Conference en Recherche d'information et Applications, CORIA 2010, Sousse, Tunisia, pp. 67-82, (ARIA) (2010)
• 2009
  • D. Buffoni, N. Usunier, P. Gallinari : “LIP6 at INEX’09: OWPC for Ad Hoc Track”, 8^th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2009, vol. 6203, Lecture Notes in Computer Science, Brisbane, Australia, pp. 59-69, (Springer) (2009)
  • N. Usunier, D. Buffoni, P. Gallinari : “Ranking with ordered weighted pairwise classification”, International Conference on Machine Learning, Montreal, Quebec, Canada, pp. 1057-1064, (ACM) (2009)
  • A. Trinh, D. Buffoni, P. Gallinari : “Probabilistic Multi-classifier by SVM from voting rule to voting features”, Extraction et gestion des connaissances (EGC'2009), vol. RNTI-E-15, Revue des Nouvelles Technologies de l'Information, Strasbourg, France, pp. 433-434 (2009)
• 2008
  • A. Trinh, D. Buffoni, P. Gallinari : “Classifieur probabiliste avec Support Vector Machine (SVMs) et Okapi”, Actes de conference TALN08, Avignon, France, pp. 75-84 (2008)