Information filtering based on transferring similarity

TL;DR

This paper introduces a new user similarity index called transferring similarity, which enhances collaborative filtering accuracy by incorporating high-order similarities and approaches optimal performance near a critical parameter value.

Contribution

The paper proposes a novel transferring similarity index and a modified collaborative filtering algorithm that significantly improves prediction accuracy over standard methods.

Findings

The new algorithm outperforms standard collaborative filtering in accuracy.

Performance peaks near a critical parameter value, indicating optimal conditions.

The method is relevant to missing link prediction problems.

Abstract

In this Brief Report, we propose a new index of user similarity, namely the transferring similarity, which involves all high-order similarities between users. Accordingly, we design a modified collaborative filtering algorithm, which provides remarkably higher accurate predictions than the standard collaborative filtering. More interestingly, we find that the algorithmic performance will approach its optimal value when the parameter, contained in the definition of transferring similarity, gets close to its critical value, before which the series expansion of transferring similarity is convergent and after which it is divergent. Our study is complementary to the one reported in [E. A. Leicht, P. Holme, and M. E. J. Newman, Phys. Rev. E {\bf 73} 026120 (2006)], and is relevant to the missing link prediction problem.