Social Trust Prediction via Max-norm Constrained 1-bit Matrix Completion

TL;DR

This paper introduces a novel max-norm constrained matrix completion method tailored for social trust prediction using 1-bit data, effectively handling non-uniform sampling and sign measurements.

Contribution

It proposes a new 1-bit max-norm constrained formulation and an efficient optimization algorithm for social trust prediction, addressing key challenges in the field.

Findings

Outperforms existing methods on benchmark datasets.

Effectively handles 1-bit and non-uniform sampling issues.

Demonstrates superior accuracy in trust prediction tasks.

Abstract

Social trust prediction addresses the significant problem of exploring interactions among users in social networks. Naturally, this problem can be formulated in the matrix completion framework, with each entry indicating the trustness or distrustness. However, there are two challenges for the social trust problem: 1) the observed data are with sign (1-bit) measurements; 2) they are typically sampled non-uniformly. Most of the previous matrix completion methods do not well handle the two issues. Motivated by the recent progress of max-norm, we propose to solve the problem with a 1-bit max-norm constrained formulation. Since max-norm is not easy to optimize, we utilize a reformulation of max-norm which facilitates an efficient projected gradient decent algorithm. We demonstrate the superiority of our formulation on two benchmark datasets.