Bibek Paudel presented a paper at the 11th ACM Conference on Recommender Systems (RecSys), a top-tier conference in computer science. The paper “Fewer Flops at the Top: Accuracy, Diversity, and Regularization in Two-Class Collaborative Filtering” introduces the problem of Two Class Collaborative Filtering (TCCF), and a novel machine learning approach to reduce the number of negative links in recommendation lists. Recommendation algorithms like this are widely used in social network and e-commerce services, to suggest new friends, jobs, books, or movies. The authors of the paper are Bibek Paudel (UZH DDIS), Thilo Haas (UZH), and Abraham Bernstein (UZH DDIS). The paper is available in the ACM digital library [1] as well as in Merlin [2], the faculty publication manager.

Bibek Paudel presenting the paper at RecSys