Information filtering via hybridization of similarity preferential diffusion processes

TL;DR

This paper introduces two similarity preferential diffusion processes for recommender systems, demonstrating improved accuracy and diversity over existing algorithms through extensive experiments, especially highlighting differences in performance between dense and sparse data systems.

Contribution

The paper proposes novel similarity preferential diffusion methods and shows they outperform current state-of-the-art recommendation algorithms, with analysis on network sparsity effects.

Findings

Improved recommendation accuracy and diversity with the new diffusion processes.

Hybridization significantly outperforms existing algorithms.

Network sparsity impacts recommendation performance and conclusions.

Abstract

The recommender system is one of the most promising ways to address the information overload problem in online systems. Based on the personal historical record, the recommender system can find interesting and relevant objects for the user within a huge information space. Many physical processes such as the mass diffusion and heat conduction have been applied to design the recommendation algorithms. The hybridization of these two algorithms has been shown to provide both accurate and diverse recommendation results. In this paper, we proposed two similarity preferential diffusion processes. Extensive experimental analyses on two benchmark data sets demonstrate that both recommendation and accuracy and diversity are improved duet to the similarity preference in the diffusion. The hybridization of the similarity preferential diffusion processes is shown to significantly outperform the state-of-art recommendation algorithm. Finally, our analysis on network sparsity show that there is significant difference between dense and sparse system, indicating that all the former conclusions on recommendation in the literature should be reexamined in sparse system.