Supercharging Recommender Systems using Taxonomies for Learning User Purchase Behavior

TL;DR

This paper introduces a taxonomy-aware latent factor model that combines human-labeled categories with latent factors to improve recommendation accuracy, especially in sparse and cold start scenarios, and demonstrates its effectiveness on large-scale shopping data.

Contribution

The paper proposes a novel taxonomy-aware latent factor model (TF) that integrates taxonomies with latent factors, along with scalable algorithms and extensions for temporal dynamics.

Findings

TF models outperform existing approaches in prediction accuracy.

TF models are scalable to large datasets with efficient algorithms.

Incorporating taxonomies improves recommendation quality in cold start and sparse data scenarios.

Abstract

Recommender systems based on latent factor models have been effectively used for understanding user interests and predicting future 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 exploit 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 (clusters) 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 develop 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 empirically evaluate the TF model for the task of predicting user purchases using a real-world shopping dataset spanning more than a million users and products. Our experiments demonstrate the benefits of using our TF models over existing approaches, in terms of both prediction accuracy and running time.