Supercharging Recommender Systems using Taxonomies for Learning User Purchase Behavior

*co-first authors Recommender systems based on latent factor models have been ef-fectively used for understanding user interests and predicting fu-ture actions. Such models work by projecting the users and items into a smaller dimensional space, thereby clustering similar users and items together and subsequently compute similarity between unknown user-item pairs. When user-item interactions are sparse (sparsity problem) or when new items continuously appear (cold start problem), these models perform poorly. In this paper, we ex-ploit the combination of taxonomies and latent factor models to mitigate these issues and improve recommendation accuracy. We observe that taxonomies provide structure similar to that of a latent factor model: namely, it imposes human-labeled categories (clus-ters) over items. This leads to our proposed taxonomy-aware latent factor model (TF) which combines taxonomies and latent factors using additive models. We develop efficient algorithms to train the TF models, which scales to large number of users/items and de-velop scalable inference/recommendation algorithms by exploiting the structure of the taxonomy. In addition, we extend the TF model to account for the temporal dynamics of user interests using high-order Markov chains. To deal with large-scale data, we develop a parallel multi-core implementation of our TF model. We empir-ically evaluate the TF model for the task of predicting user pur-chases using a real-world shopping dataset spanning more than a million users and products. Our experiments demonstrate the ben-efits of using our TF models over existing approaches, in terms of both prediction accuracy and running time. 1