Many modern recommender systems rely on matrix factor-ization techniques to produce personalized recommendationson the basis of the feedback that users provided on differ-ent items in the past. The feedback may take different forms,such as the rating of a movie, or the number of times a userlistened to the songs of a given music band. Nonetheless, insome situations, the user can perform several actions on eachitem, and the feedback is multidimensional (e.g., the user ofan e-commerce website can either click on a product, add theproduct to her cart or buy it). In this case, one can no longerview the recommendation problem as a matrix completion,unless the problem is reduced to a series of multiple inde-pendent problems, thus loosing the correlation between thedifferent actions. In this case, the most suitable approach is touse a tensor approach to learn all dimensions of the feedbacksimultaneously. In this paper, we propose a specific instanceof tensor completion and we show how it can be heavily par-allelized over both the dimensions (i.e., items, users, actions)and within each dimension (i.e., each item separately). Wevalidate the proposed method both in terms of prediction ac-curacy and scalability to large datasets.
Single and coupled matrix-tensor factorization (CMTF) has been widely used in this context for imputation-based recommendation from ratings, social network, and other user-item data. When this side information is in the form of item-item correlation matrices or graphs, existing CMTF algorithms may fall short. Alleviating current limitations, we introduce a novel model coined coupled graph-tensor factorization (CGTF) that judiciously accounts for graph-related side information. The CGTF model has the potential to overcome practical challenges, such as missing slabs from the tensor and/or missing rows/columns from the correlation matrices. A novel alternating direction method of multipliers (ADMM) is also developed that recovers the nonnegative factors of CGTF. Our algorithm enjoys closed-form updates that result in reduced computational complexity and allow for convergence claims. A novel direction is further explored by employing the interpretable factors to detect graph communities having the tensor as side information. The resulting community detection approach is successful even when some links in the graphs are missing. Results with real data sets corroborate the merits of the proposed methods relative to state-of-the-art competing factorization techniques in providing recommendations and detecting communities.
Recommender systems have been widely adopted by electronic commerce and entertainment industries for individualized prediction and recommendation, which benefit consumers and improve business intelligence. In this article, we propose an innovative method, namely the recommendation engine of multilayers (REM), for tensor recommender systems. The proposed method utilizes the structure of a tensor response to integrate information from multiple modes, and creates an additional layer of nested latent factors to accommodate between-subjects dependency. One major advantage is that the proposed method is able to address the "cold-start" issue in the absence of information from new customers, new products or new contexts. Specifically, it provides more effective recommendations through sub-group information. To achieve scalable computation, we develop a new algorithm for the proposed method, which incorporates a maximum block improvement strategy into the cyclic blockwise-coordinate-descent algorithm. In theory, we investigate both algorithmic properties for global and local convergence, along with the asymptotic consistency of estimated parameters. Finally, the proposed method is applied in simulations and IRI marketing data with 116 million observations of product sales. Numerical studies demonstrate that the proposed method outperforms existing competitors in the literature.
Wu, Fei (Zhejiang University) | Tan, Xu (Zhejiang University) | Yang, Yi (University of Queensland) | Tao, Dacheng (University of Technology, Sydney) | Tang, Siliang (Zhejiang University) | Zhuang, Yueting (Zhejiang University)
Non-negative tensor factorization (NTF) has attracted great attention in the machine learning community. In this paper, we extend traditional non-negative tensor factorization into a supervised discriminative decomposition, referred as Supervised Non-negative Tensor Factorization with Maximum-Margin Constraint(SNTFM2). SNTFM2 formulates the optimal discriminative factorization of non-negative tensorial data as a coupled least-squares optimization problem via a maximum-margin method. As a result, SNTFM2 not only faithfully approximates the tensorial data by additive combinations of the basis, but also obtains a strong generalization power to discriminative analysis (in particularfor classification in this paper). The experimental results show the superiority of our proposed model over state-of-the-art techniques on both toy and real world data sets.
About this course: In this course you will learn a variety of matrix factorization and hybrid machine learning techniques for recommender systems. Starting with basic matrix factorization, you will understand both the intuition and the practical details of building recommender systems based on reducing the dimensionality of the user-product preference space. Then you will learn about techniques that combine the strengths of different algorithms into powerful hybrid recommenders.