PD-SRS: personalized diversity for a fair session-based recommendation system

Naime Ranjbar Kermany*, Luiz Pizzato, Jian Yang, Shan Xue, Jia Wu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

Abstract

Session-based Recommender Systems (SRSs), which aim to recommend users’ next action based on their current and historical sessions, play a significant role in many real-world online services. The existing session-based recommendation methods have mainly focused on the accuracy of recommendation, which biases to reinforce popular items/services and loses the recommendation diversity. Diversity is a positive aspect particularly in SRSs as the target user may like to be surprised and interact with a broader range of content in different sessions. In this work, we propose a Personalized Diversification strategy for a Session-based Recommender System (PD-SRS) using graph neural networks. Comprehensive experiments are carried out on two real-world datasets to demonstrate the effectiveness of PD-SRS in making a trade-off between accuracy and personalized diversity over the baselines.

Original languageEnglish
Title of host publicationService-oriented computing
Subtitle of host publication20th International Conference, ICSOC 2022, Seville, Spain, November 29 – December 2, 2022, proceedings
EditorsJavier Troya, Brahim Medjahed, Mario Piattini, Lina Yao, Pablo Fernández, Antonio Ruiz-Cortés
Place of PublicationCham, Switzerland
PublisherSpringer, Springer Nature
Pages331-339
Number of pages9
ISBN (Electronic)9783031209840
ISBN (Print)9783031209833
DOIs
Publication statusPublished - 2022
Event20th International Conference on Service-Oriented Computing, ICSOC 2022 - Seville, Spain
Duration: 29 Nov 20222 Dec 2022

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume13740
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference20th International Conference on Service-Oriented Computing, ICSOC 2022
Country/TerritorySpain
CitySeville
Period29/11/222/12/22

Keywords

  • Session-based recommendation
  • Personalized diversity
  • Fairness
  • Long-tail recommendation
  • Graph neural network

Fingerprint

Dive into the research topics of 'PD-SRS: personalized diversity for a fair session-based recommendation system'. Together they form a unique fingerprint.

Cite this