Two-Way Grouping by One-Way Topic Models

Reference:

Eerika Savia, Kai Puolamäki, and Samuel Kaski. Two-way grouping by one-way topic models. In N. Adams, C. Robardet, A. Siebes, and J.-F. Boulicaut, editors, Advances in Intelligent Data Analysis VIII, Proceedings of the 8th International Symposium on Intelligent Data Analysis, IDA 2009, Lecture Notes in Computer Science, pages 178–189, Berlin, 2009. Springer.

Abstract:

We tackle the problem of new users or documents in collaborative filtering. Generalization over users by grouping them into user groups is beneficial when a rating is to be predicted for a relatively new document having only few observed ratings. The same applies for documents in the case of new users. We have shown earlier that if there are both new users and new documents, two-way generalization becomes necessary, and introduced a probabilistic Two-Way Model for the task. The task of finding a two-way grouping is a non-trivial combinatorial problem, which makes it computationally difficult. We suggest approximating the Two-Way Model with two URP models; one that groups users and one that groups documents. Their two predictions are combined using a product of experts model. This combination of two one-way models achieves even better prediction performance than the original Two-Way Model.

Suggested BibTeX entry:

@inproceedings{Savia09ida,
    address = {Berlin},
    author = {Eerika Savia and Kai Puolam{\"a}ki and Samuel Kaski},
    booktitle = {Advances in Intelligent Data Analysis VIII, Proceedings of the 8th International Symposium on Intelligent Data Analysis, IDA 2009},
    editor = {N. Adams and C. Robardet and A. Siebes and J.-F. Boulicaut},
    pages = {178--189},
    publisher = {Springer},
    series = {Lecture Notes in Computer Science},
    title = {Two-Way Grouping by One-Way Topic Models},
    year = {2009},
}