Synthetic sequence generator for recommender systems - memory biased random walk on sequence multilayer network

TL;DR

This paper introduces a memory biased random walk model on multilayer sequence networks to generate synthetic data, enabling recommender system training while respecting user privacy restrictions.

Contribution

It proposes a novel synthetic data generator based on memory biased random walks on multilayer networks for privacy-preserving recommender system training.

Findings

Synthetic data improves recommender system performance.

Model maintains privacy while providing useful training data.

Applicable to privacy-sensitive environments.

Abstract

Personalized recommender systems rely on each user's personal usage data in the system, in order to assist in decision making. However, privacy policies protecting users' rights prevent these highly personal data from being publicly available to a wider researcher audience. In this work, we propose a memory biased random walk model on multilayer sequence network, as a generator of synthetic sequential data for recommender systems. We demonstrate the applicability of the synthetic data in training recommender system models for cases when privacy policies restrict clickstream publishing.