Scalable Recommender Systems through Recursive Evidence Chains
This work addresses scalability and cold-start issues in recommender systems for large-scale applications, though it appears incremental as it builds on existing matrix factorization methods.
The paper tackled the scalability problem in recommender systems by developing a novel approach that generates latent variables on demand from the ratings matrix and a fixed parameter pool, using chains of evidence to link missing ratings to prototypical users and items, resulting in competitive accuracy and convergence speed compared to current matrix factorization techniques.
Recommender systems can be formulated as a matrix completion problem, predicting ratings from user and item parameter vectors. Optimizing these parameters by subsampling data becomes difficult as the number of users and items grows. We develop a novel approach to generate all latent variables on demand from the ratings matrix itself and a fixed pool of parameters. We estimate missing ratings using chains of evidence that link them to a small set of prototypical users and items. Our model automatically addresses the cold-start and online learning problems by combining information across both users and items. We investigate the scaling behavior of this model, and demonstrate competitive results with respect to current matrix factorization techniques in terms of accuracy and convergence speed.