2016/12/17 by Mirko Polato, Polato, Mirko, Fabio Aiolli +1
Computer Science · Decision Sciences · Engineering · #Advanced Bandit Algorithms Research #Advanced MIMO Systems Optimization #Artificial Intelligence (cs.AI) #Expert finding and Q&A systems #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Machine Learning (cs.LG) #Recommender Systems and Techniques
paper · pdf · doi:10.48550/arxiv.1612.05729
openalex publication_date 2016/12/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The increasing availability of implicit feedback datasets has raised the\ninterest in developing effective collaborative filtering techniques able to\ndeal asymmetrically with unambiguous positive feedback and ambiguous negative\nfeedback. In this paper, we propose a principled kernel-based collaborative\nfiltering method for top-N item recommendation with implicit feedback. We\npresent an efficient implementation using the linear kernel, and we show how to\ngeneralize it to kernels of the dot product family preserving the efficiency.\nWe also investigate on the elements which influence the sparsity of a standard\ncosine kernel. This analysis shows that the sparsity of the kernel strongly\ndepends on the properties of the dataset, in particular on the long tail\ndistribution. We compare our method with state-of-the-art algorithms achieving\ngood results both in terms of efficiency and effectiveness.\n