Transferring user interests across websites with unstructured text for cold-start recommendation

Abstract

In this work, we investigate the possibility of cross-website transfer learning for tackling the cold-start problem. To address the cold-start issues commonly present in a collaborative filtering (CF) system, most existing cross-domain CF models require auxiliary rating data from another domain; nevertheless, under the cross-website scenario, such data is often unobtainable. Therefore, we propose the nearest-neighbor transfer matrix factorization (NT-MF) model, where a topic model is applied to the unstructured user-generated content in the source domain, and the similarity between users in the latent topic space is utilized to guide the target-domain CF model. Specifically, the latent factors of the nearest-neighbors are regarded as a set of pseudo observations, which can be used to estimate the unknown parameters in the model. Improvement over previous methods, especially for the cold-start users, is demonstrated with experiments on a real-world cross-website dataset.