Synthetic Sequence Generator for Recommender Systems – Memory Biased Random Walk on a Sequence Multilayer Network
Nino Antulov-Fantulina✉, Matko Bosnjakb, V. Zlatićc, Miha Grcard, T. Šmuca
- aLaboratory for Information Systems, Division of Electronics, Rudjer Bošković Institute, Zagreb, Croatia
- bDepartment of Computer Science, University College London, London, UK
- cTheoretical Physics Division, Rudjer Bošković Institute, Zagreb, Croatia
- dDepartment of Knowledge Technologies - E8, Jožef Stefan Institute, Ljubljana, Slovenia
- ✉Corresponding author
Papers open inside SciRah from the original publisher or repository; SciRah never hosts them.
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 a multilayer sequence network, as a generator of synthetic sequential data for recommender systems. We demonstrate the applicability of the generated synthetic data in training recommender system models in cases when privacy policies restrict clickstream publishing.
TL;DRThis work proposes a memory biased random walk model on a multilayer sequence network, as a generator of synthetic sequential data for recommender systems, and demonstrates the applicability of the generated synthetic data in training recommender system models in cases when privacy policies restrict clickstream publishing. (Semantic Scholar)