Recommender Systems present a high-level of sparsity in their ratings matrices. The collaborative filtering sparse data makes it difficult to: 1) compare elements using memory-based solutions; 2) obtain precise models using model-based solutions; 3) get accurate predictions; and 4) properly cluster elements. We propose the use of a Bayesian non-negative matrix factorization (BNMF) method to improve the current clustering results in the collaborative filtering area. We also provide an original pre-clustering algorithm adapted to the proposed probabilistic method. Results obtained using several open data sets show: 1) a conclusive clustering quality improvement when BNMF is used, compared with the classical matrix factorization or to the improved KMeans results; 2) a higher predictions accuracy using matrix factorizationbased methods than using improved KMeans; and 3) better BNMF execution times compared with those of the classic matrix factorization, and an additional improvement when using the proposed pre-clustering algorithm.
The existing scientific documentation-based recommender systems focus on exploiting the citations and references information included in each research paper and also the lists of co-authors. In this way, it can be addressed the recommendation of related papers and even related authors. The approach we propose is original because instead of using each paper citations and co-authors, we relate each of the papers with their main research topics. This approach provides a semantic level superior to that currently used, which allows us to obtain useful results. We can use collaborative filtering recommender systems to recommend research topics related to each paper and also to recommend papers related to each research topic. In order to face this innovative proposal, we have solved a series of challenges that allow us to offer various resources and results in the paper. Our main contributions are: 1) making a data mining of scientific documentation; 2) creating and publishing an open database containing the data mining results; 3) extracting the research topics from the available scientific documentation; 4) creating and publishing a recommender system data set obtained from the database and the research topics; 5) testing the data set through a complete set of collaborative filtering methods and quality measures; and 6) selecting and showing the best methods and results, obtained using the open data set, in the context of scientific documentation recommendations. Results of the paper show the suitability of the provided data set in collaborative filtering processes, as well as the superiority of the model-based methods to face scientific documentation recommendations.
In this paper, we propose a collaborative filtering method designed to improve the current memory-based prediction times without worsening and even improving the existing accuracy results. The accuracy improvement is achieved by combining the numerical relevance of the ratings with non-numerical information based on the votes structure. The improvement of the prediction time is achieved by setting four actions: 1) simplification of the similarity measure design, in order to minimize the necessary calculations; 2) construction and maintenance of a model that simplifies the predictions processing; 3) optimization of the computation, using a set-based model and a bit-based processing implementation; and 4) switching between the bit processing and the numerical processing, depending on the density of the users' ratings. Experimental results show the improvements both in the prediction time and the accuracy. Experiments have used a significant amount of state-of-the-art baselines and collaborative filtering public data sets.
Recommendation to group of users is a big challenge for collaborative filtering. The recommendations to groups of users arise from the convenience of being able to recommend a group of users about products or services that satisfy the entire group. In this paper, we propose the similarity measure SMGU, tailored for collaborative filtering recommendations to groups of users. This similarity measure combines both numerical and non-numerical information. Numerical information is weighted attending to the rating singularity of the group members. This paper focuses on the assumption that the singularity of the ratings cast by the users of the group is a relevant information for finding suitable neighbors. For each item, we consider that a rating is singular for a group or for a user when that rating is different from the majority of the rating cast by the other users. Non-numerical structural information can be considered as valuable to match group preferences with neighbors preferences. Experiments have been run using open recommender systems datasets. Compared with representative baselines, results show accuracy improvements when the proposed method is used. Additionally, this paper provides a section devoted to the experiments reproducibility issue. Finally, this work opens opportunities to face new challenges in the recommendation to group of users: explanation of recommendations, determination of reliability measures, and improvement of accuracy, novelty and diversity results.