Supervision : Patrick GALLINARI

Co-supervision : GUIGUE Vincent

Representation Learning of User-generated Data

In this thesis, we study how representation learning methods can be applied to user-generated data. Our contributions cover three different applications but share a common denominator: the extraction of relevant user representations. Our first application is the item recommendation task, where recommender systems build user and item profiles out of past ratings reflecting user preferences and item characteristics. Nowadays, textual information is often together with ratings available and we propose to use it to enrich the profiles extracted from the ratings. Our hope is to extract from the textual content shared opinions and preferences. The models we propose provide another opportunity: predicting the text a user would write on an item. Our second application is sentiment analysis and, in particular, polarity classification. Our idea is that recommender systems can be used for such a task. Recommender systems and traditional polarity classifiers operate on different time scales. We propose two hybridizations of these models: the former has better classification performance, the latter highlights a vocabulary of surprise in the texts of the reviews. The third and final application we consider is urban mobility. It takes place beyond the frontiers of the Internet, in the physical world. Using authentication logs of the subway users, logging the time and station at which users take the subway, we show that it is possible to extract robust temporal profiles.

2014-2016 Publications

• 2016
  • M. Poussevin, E. Tonnelier, N. Baskiotis, V. Guigue, P. Gallinari : “Mining ticketing logs for usage characterization with nonnegative matrix factorization”, chapter in Big Data Analytics in the Social and Ubiquitous Context, vol. 9546, Lecture Notes in Computer Science, pp. 147-164, (Springer), (ISBN: 978-3-319-29008-9) (2016)
• 2015
  • M. Poussevin : “Apprentissage de représentations pour des données générés par les utilisateurs”, thesis, phd defence 01/21/2015, supervision Gallinari, Patrick, co-supervision : Guigue, Vincent (2015)
  • M. Poussevin, V. Guigue, P. Gallinari : “Extended Recommendation Framework: Generating the Text of a User Review as a Personalized Summary”, 2^nd Workshop on New Trends in Content-Based Recommender Systems, ACM RecSys, vol. 1448, CEUR Workshop Proceedings, Vienna, Austria, pp. 34-41, (CEUR-WS.org) (2015)
  • M. Poussevin, V. Guigue, P. Gallinari : “Extraction d’un vocabulaire de surprise par mélange de filtrage collaboratif et d’analyse de sentiments.”, CORIA 2015 - 12^e Conférence en Recherche d'Informations et Applications, Paris, France, pp. 123-138 (2015)
• 2014
  • M. Poussevin, V. Guigue, P. Gallinari : “Extended recommendation framework: Generating the text of a user review as a personalized summary”, (2014)
  • M. Poussevin, N. Baskiotis, V. Guigue, P. Gallinari : “Mining ticketing logs for usage characterization with nonnegative matrix factorization”, SenseML 2014 -- ECML Workshop, Nancy, France (2014)
  • M. Poussevin, N. Baskiotis, V. Guigue, P. Gallinari : “Factorisation matricielle sous contraintes pour l’analyse des usages du métro parisien”, CAp'2014 : Conférence d'Apprentissage Automatique, Saint-Etienne, France (2014)
  • M. Poussevin, E. Guardia‑Sebaoun, V. Guigue, P. Gallinari : “Recommandation par combinaison de filtrage collaboratif et d’analyse de sentiments”, CORIA 2014 - COnférence en Recherche d’Information et Applications, Nancy, France, pp. 27-42 (2014)