Session-based Recommendations with Recurrent Neural Networks
This addresses the challenge for real-life recommender systems, such as small e-commerce websites, that rely on limited session data instead of long user histories, offering a more accurate alternative to item-to-item recommendations.
The paper tackles the problem of making accurate recommendations from short session-based data, where traditional matrix factorization fails, by proposing an RNN-based approach that models entire sessions; experimental results on two datasets show marked improvements over widely used methods.
We apply recurrent neural networks (RNN) on a new domain, namely recommender systems. Real-life recommender systems often face the problem of having to base recommendations only on short session-based data (e.g. a small sportsware website) instead of long user histories (as in the case of Netflix). In this situation the frequently praised matrix factorization approaches are not accurate. This problem is usually overcome in practice by resorting to item-to-item recommendations, i.e. recommending similar items. We argue that by modeling the whole session, more accurate recommendations can be provided. We therefore propose an RNN-based approach for session-based recommendations. Our approach also considers practical aspects of the task and introduces several modifications to classic RNNs such as a ranking loss function that make it more viable for this specific problem. Experimental results on two data-sets show marked improvements over widely used approaches.