Attack Detection Using Item Vector Shift in Matrix Factorisation Recommenders

doi:10.1145/3721285

Attack Detection Using Item Vector Shift in Matrix Factorisation Recommenders

2025-02-28

0

OA

AI

DOI：10.1145/3721285

Original request for help

Save

Abstract

En 中文

This article proposes a novel method for detecting shilling attacks in Matrix Factorization (MF)-based Recommender Systems (RSs), in which attackers use false user-item feedback to promote a specific item. Unlike existing methods that use either supervised learning to distinguish between attack and genuine profiles or analyze target item rating distributions to detect false ratings, our method uses an unsupervised technique to detect false ratings by examining shifts in item preference vectors that exploit rating deviations and user characteristics, making it a promising new direction. The experimental results demonstrate the effectiveness of our approach in various attack scenarios, including those involving obfuscation techniques.

Keywords:

Data poisoning attacks

matrix factorization

recommender systems

at-tack detection

Attack Detection Using Item Vector Shift in Matrix Factorisation Recommenders

Abstract

Journal

Organization