The work developed at the LATIN - Laboratory for Treating Information - has covered some of the key areas in modern IR from crawling, indexing, compression and ranking methods to search engines and recommender systems. Further, its focus on addressing practical problems of relevance to society and on building prototypes to validate the proposed solutions has led to the spin-off of two key start-up companies in Brazil, one of them acquired by Google Inc. to become its R&D center for Latin America.
With more than seventy graduate students formed, many of them occupying key positions in academia and industry, and a large scientific production spread across many of the major IR journals and conferences, the group has established a solid reputation as a world class research group in its topics of interest. Today, the group is focused on exploring technologies related to web IR and their application to practical problems—a trademark of the group over the years.
A. Bessa, R. L. T. Santos, A. Veloso, N. Ziviani, Exploiting Item Co-Utility to Improve Collaborative Filtering Recommendations. (to appear in) JASIST, 2017.
F. Moraes, R. L. T. Santos, N. Ziviani, UFMG at the TREC 2016 Dynamic Domain track. TREC, 2016.
R. Lopes, R. Assunção, R. L. T. Santos, Efficient Bayesian Methods for Graph-based Recommendation. RecSys, 333-340, 2016.