Graph Neural Networks in Recommendation Systems for Superior User Experiences

Title

Graph Neural Networks in Recommendation Systems for Superior User Experiences

Creator

Upadhyay, Priyansha; Banu, P.K. Nizar

Description

Recommender systems are now an essential tool for traversing huge amounts of online information, especially as user tastes change in dynamic settings. They can be classified into different types based on their use, such as collaborative filtering and content-based recommendation. Graph Neural Networks (GNNs), however, are best suited to learn from graph-structured data and have become a revolutionary technology in building recommendation systems based on their capacity to capture intricate relationships and dependencies in graph-structured data. Unlike traditional methods, GNN-based recommendation systems capture both local and global connectivity patterns in user-item interaction graphs, enhancing prediction accuracy and robustness. This study looks at key types of Graph Neural Networks, known as GNNs, and includes Graph Convolutional Networks (GCNs), Graph Attention Networks (GATs), and GraphSAGE. Each type is good at tackling specific challenges like making personalized recommendations, handling cold-start problems, and managing large-scale data effectively. Big companies across different industries are using GNNs to enhance their services. For example, e-commerce giants like Amazon and Alibaba, and streaming platforms like Netflix and Spotify, use these networks to boost user engagement and satisfaction. The study also covers strategies for applying GNNs. This includes constructing user-item bipartite graphs, including domain-specific features, and employing contrastive learning methods to improve functionality. By combining theoretical concepts with real-world usage, we hope to offer an informed view of GNNs in recommendation systems. In addition, metrics like Precision, Recall, and Normalized Discounted Cumulative Gain (nDCG), providing implementable recommendations for future development in this fast-moving field, will be addressed. 2026 by John Wiley & Sons Inc. All rights reserved.

Source

Next-Generation Recommendation Systems: A Comprehensive Guide to Enabling Technologies and Tools and their Business Benefits;pp.121-150

Date

01-01-2026

Publisher

wiley

Subject

Graph Attention Networks (GATs); Graph Convolutional Networks (GCNs); Graph Neural Networks (GNNs); GraphSAGE; Recommendation systems

Coverage

Upadhyay P., Department of Computer Science, Christ University, Bangalore, India; Banu P.K.N., Department of Computer Science, Christ University, Bangalore, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISBN: 978-139435157-2; 978-139435154-1;

Format

online

Language

English

Type

Book chapter

Identifier

https://doi.org/10.1002/9781394351572.ch6

https://www.scopus.com/pages/publications/105038490822?origin=resultslist