Improving tensor based recommenders with clustering

Intelligent Web and Information Systems - Machine Intelligence, Computer Science, Aalborg University
Martin Leginus Peter Dolog and Valdas Zemaitis

Detailed results of experiments for each clustering technique, with different similarity measures for Bibsonomy dataset. Files contain detailed list of generated recommendations for all users.

Detailed results of experiments for Correlated Feature Hashing using Dice similarity measure for Movielens dataset. Files contain detailed list of generated recommendations for all users.

Implemented recommendation system - source code. Please, import extracted files into Eclipse IDE and then run ApplicationStarter.java.