Highly accurate recommendation algorithm based on high-order similarities

TL;DR

This paper presents a modified collaborative filtering algorithm that uses diffusion-based similarity and second order considerations to significantly improve recommendation accuracy, diversity, and popularity control.

Contribution

The authors introduce a novel MCF algorithm utilizing diffusion-based user similarities and second order similarities to enhance recommendation performance.

Findings

Ranking score improved by 24.9%

Outperforms standard collaborative filtering in accuracy

Balances diversity and popularity effectively

Abstract

In this Letter, we introduce a modified collaborative filtering (MCF) algorithm, which has remarkably higher accuracy than the standard collaborative filtering. In the MCF, instead of the standard Pearson coefficient, the user-user similarities are obtained by a diffusion process. Furthermore, by considering the second order similarities, we design an effective algorithm that depresses the influence of mainstream preferences. The corresponding algorithmic accuracy, measured by the ranking score, is further improved by 24.9% in the optimal case. In addition, two significant criteria of algorithmic performance, diversity and popularity, are also taken into account. Numerical results show that the algorithm based on second order similarity can outperform the MCF simultaneously in all three criteria.