RIM3@UAM Information Retrieval on different media based on multidimensional models: relevance, novelty, personalization and context

Project summary

Funded from January 2009 to December 2011 by the Spanish Ministry of Science and Education (TIN2008-06566-C04-02), RIM3@UAM is part of the RIM3 coordinated project, in which the Autónoma University of Madrid (UAM) participates jointly with University of Granada (coordinator) and University of A Coruña.

The RIM3 project brings together complementary research lines which tackle the Information Retrieval (IR) problem from a multidimensional perspective. The transcendence of IR across media, contexts and modalities is a fact today. While information sources, media, and delivery paradigms proliferate, user information needs highly depend on factors such as time, place, and history of interaction, task at hand, current user interests, etc. To meet the challenges raised in this evolution, IR models that go beyond the usual relevance-oriented approach need to be developed for IR technologies to keep up with the deep transformations in the retrieval spaces, businesses, users' demands, and how IR problems themselves are understood. Besides the dimension of relevance, we study in this project how to endow the systems with advanced capabilities for novelty detection, redundancy filtering, subtopic detection, personalization and context-based retrieval, opening new ways to improve the quality of access to information sources. These dimensions are not only considered for the basic retrieval task but also for other tasks such as automatic summarization, document clustering and categorization.

The UAM subproject focuses on different topics in the spectrum of personalized IR, such as context modeling for IR, personalized search, hybrid recommender systems and rank fusion. Research on context-aware strategies focuses on the analysis of live user activities, and their interrelation to historic, long-term information as is generally used to build stable user profiles, as a means to devise enhanced, more effective personalized IR techniques. In the area of recommender systems, our research aims to overcome the limitations of data sparsity by discovering hidden links in the data, involving users, information items, users' actions, inputs, tags, and so forth, involved in recommendation problems. The heterogeneity of studied aproaches in the project motivates the research of dynamic hibridization techniques that select and combine retrieval strategies in optimal ways depending on the situation. Problems involving high levels of uncertainty as is the case of personalization based on implicit user input, with high and fast variability in the contextual conditions, indeed call for hybrid solutions, which generally provide the best results in such cases. Our aproach in this project delves into dynamic performance prediction methods as a means to select or prioritize the available retrieval methods according to the expected quality of their outputs.

Project participants


Project-related publications

I. Cantador and P. Castells. Group Recommender Systems: New Perspectives in the Social Web. In J. J. Pazos Arias, A. Fernández Vilas, and R. P. Díaz Redondo (Eds.), Recommender Systems for the Social Web. Springer Verlag Intelligent Systems Reference Library Vol. 32, ISBN 978-3-642-25693-6, 2012.

S. Vargas and P. Castells. Rank and Relevance in Novelty and Diversity Metrics for Recommender Systems. 5th ACM International Conference on Recommender Systems (RecSys 2011). Chicago, Illinois, October 2011.
PDF

A. Bellogín, P. Castells, and I. Cantador. Precision-Based Evaluation of Recommender Systems: An Algorithmic Comparison. 5th ACM International Conference on Recommender Systems (RecSys 2011). Chicago, Illinois, October 2011.
PDF

A. Bellogín, J. Wang, and P. Castells. Structured Collaborative Filtering. 20th ACM Conference on Information and Knowledge Management (CIKM 2011), Poster Session. Glasgow, UK, October 2011.
PDF

A. Bellogín, P. Castells, and I. Cantador. Predicting the Performance of Recommender Systems: An Information Theoretic Approach. 3rd International Conference on the Theory of Information Retrieval (ICTIR 2011). Bertinoro, Italy, September 2011. Springer Verlag Lecture Notes in Computer Science.
PDF

I. Cantador and P. Castells. Extracting Multilayered Communities of Interest from Semantic User Profiles: Application to Group Modeling and Hybrid Recommendations. Computers in Human Behavior 27, special issue on Social and Humanistic Computing for the Knowledge Society, July 2011, pp. 1321-1336.
Draft version: PDF

A. Bellogín, P. Castells and I. Cantador. Self-adjusting Hybrid Recommenders Based on Social Network Analysis. 34th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2011), Poster Session. Beijing, China, July 2011.
PDF

D. Vallet and P. Castells. On Diversifying and Personalizing Web Search. 34th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2011), Poster Session. Beijing, China, July 2011.
PDF

S. Vargas, P. Castells and D. Vallet. Intent-Oriented Diversity in Recommender Systems. 34th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2011), Poster Session. Beijing, China, July 2011.
Paper: PDF
Poster: PDF

I. Fernández-Tobías, I. Cantador, and A. Bellogín. cTag: Semantic Contextualisation of Social Tags. International Workshop on Semantic Adaptive Social Web (SASWeb 2011), at 19th User Modeling, Adaptation and Personalization (UMAP 2011). Girona, Spain, July 2011. In press.
PDF

A. Bellogín. Performance Prediction in Recommender Systems. 19th User Modeling, Adaptation and Personalization (UMAP 2011), Doctoral Consortium. Girona, Spain, July 2011. In press.
PDF

D. Vallet, M. Halvey, J. Jose and P. Castells. Applying Soft Links to Diversify Video Recommendations. 9th International Workshop on Content-Based Multimedia Indexing (CBMI 2011). Madrid, Spain, June 2011.

D. Vallet, F. Hopfgartner, J. M. Jose, and P. Castells. Effects of Usage based Feedback on Video Retrieval: a Simulation based Study. ACM Transactions on Information Systems 29(2), April 2011.
Draft version: PDF

P. Castells, S. Vargas, and J. Wang. Novelty and Diversity Metrics for Recommender Systems: Choice, Discovery and Relevance. International Workshop on Diversity in Document Retrieval (DDR 2011) at the 33rd European Conference on Information Retrieval (ECIR 2011). Dublin, Ireland, April 2011.
Paper: PDF
Poster: PDF

A. Bellogín, J. Wang, and P. Castells. Text Retrieval Methods for Item Ranking in Collaborative Filtering. 33rd European Conference on Information Retrieval (ECIR 2011). Dublin, Ireland, April 2011. Springer Verlag Lecture Notes in Computer Science.
PDF

A. Bellogín, I. Cantador, P. Castells, and Á. Ortigosa. Discerning Relevant Model Features in a Content-based Collaborative Recommender System. In J. Fürnkranz and E. Hüllermeier (Eds.), Preference Learning. Springer Verlag, ISBN 978-3-642-14124-9, 2011, pp. 429-456.
Available from: http://www.springer.com/computer/ai/book/978-3-642-14124-9

A. Bellogín, I. Cantador, and P. Castells. A Study of Heterogeneity in Recommendations for a Social Music Service. Workshop on Information Heterogeneity and Fusion in Recommender Systems (HetRec 2010), at the 4th ACM Conference on Recommender Systems (RecSys 2010). Barcelona, Spain, September 2010, pp. 1-8.
PDF

F. Díez, J. E. Chavarriaga, P. G. Campos, and A. Bellogín. Movie Recommendations Based on Explicit and Implicit Features Extracted from the Filmtipset Dataset. Workshop on Context-Aware Movie Recommendation (CAMRa 2010), at the 4th ACM Conference on Recommender Systems (RecSys 2010). Barcelona, Spain, September 2010, pp. 42-52.
PDF

P. G. Campos, A. Bellogín, F. Díez, and J. E. Chavarriaga. Simple Time-biased KNN-based Recommendations. Workshop on Context-Aware Movie Recommendation (CAMRa 2010), at the 4th ACM Conference on Recommender Systems (RecSys 2010). Barcelona, Spain, September 2010, pp. 20-23.
PDF

J. Conde, D. Vallet, P. Castells. Inferring User Intent in Web Search by Exploiting Social Annotations. 33rd Annual International ACM Conference on Research and Development in Information Retrieval (SIGIR 2010), Poster Session. Geneva, Switzerland, July 2010.
PDF

S. López, F. Díez. Sobre la distribución de scores de documentos relevantes y no-relevantes: un problema abierto. I Congreso Español de Recuperación de Información. Madrid, junio 2010.
PDF

P. G. Campos, F. Díez. La Temporalidad en los Sistemas de Recomendación: Una revisión actualizada de propuestas teóricas. I Congreso Español de Recuperación de Información. Madrid, junio 2010.
PDF

J. Conde, D. Vallet, P. Castells. Contextualización de Búsqueda Web Mediante el Uso de Anotaciones Sociales. I Congreso Español de Recuperación de Información. Madrid, junio 2010.
PDF

D. Vallet, I. Cantador, and J. M. Jose. Personalizing Web Search with Folksonomy-based User and Document Profiles. 32nd European Conference on Information Retrieval (ECIR 2010). Milton Keynes, UK, March 2010. Springer Verlag Lecture Notes in Computer Science, Vol. 5993, ISBN 978-3642122743.
PDF

A. Bellogín and P. Castells. A Performance Prediction Approach to Enhance Collaborative Filtering Performance. 32nd European Conference on Information Retrieval (ECIR 2010). Milton Keynes, UK, March 2010. Springer Verlag Lecture Notes in Computer Science, Vol. 5993, ISBN 978-3642122743, pp. 382-393.
PDF

D. Vallet, I. Cantador, and J. M. Jose. Exploiting External Knowledge to Improve Video Retrieval. 11th ACM SIGMM International Conference on Multimedia Information Retrieval (MIR 2010). Philadelphia, Pensylvania, US, March 2010.
PDF

D. Vallet, M. Halvey, D. Hannah, and J. M. Jose. A Multi Faceted Recommendation Approach for Explorative Video Retrieval Tasks. 14th ACM International Conference on Intelligent User Interfaces (IUI 2010). Hong Kong, China, February 2010.
PDF

A. Bellogín and P. Castells. Predicting Neighbor Goodness in Collaborative Filtering. 8th International Conference on Flexible Query Answering Systems (FQAS 2009). Roskilde, Denmark, October 2009. Springer Verlag Lecture Notes in Computer Science, Vol. 5822, ISBN 978-3-642-04956-9, pp. 605-616. 
PDF

I. Cantador and P. Castells. Semantic Contextualisation in a News Recommender System. Workshop on Context-Aware Recommender Systems (CARS 2009) at the 3rd ACM Conference on Recommender Systems (RecSys 2009). New York, October 2009.
PDF