Context-aware Multi-stakeholder Recommender Systems
DocUID: 2022-007 Full Text:
PDF Author: Tahereh Arabghalizi, Alexandros Labrinidis
Abstract: Traditional recommender systems help users find the most relevant products or services to match their needs and preferences. However, they overlook the preferences of other sides of the market (aka stakeholders) involved in the system. In this paper, we propose to use contextual bandit algorithms in multi-stakeholder platforms where a multi-sided relevance function with adjusting weights is modeled to consider the preferences of all involved stakeholders. This algorithm sequentially recommends the items based on the contextual features of users along with the priority of the stakeholders and their relevance to the items. Our extensive experimental results on a dataset consisting of MovieLens (1m), IMDB (81k+), and a synthetic dataset show that our proposed approach outperforms the baseline methods and provides a good trade-off between the satisfaction of different stakeholders over time.
Keywords: Recommender Systems; Multi-stakeholder Recommender Systems; Multi-armed Bandits; Contextual Bandits;
Published In: The International FLAIRS Conference Proceedings
Pages: 35
Year Published: 2022
DOI: https://doi.org/10.32473/flairs.v35i.130573
Project: PittSmartLiving Subject Area: Multi-armed Bandits, Recommender Systems
Publication Type: Conference Paper
Sponsor: NSF CNS-1739413
Similar Publications: