Grocery Recommendations Using Graph Neural Networks and Transformer Models

DOI

10.2991/978-94-6463-866-0_83How to use a DOI?

Keywords

Machine Learning; Grocery Recommendations; Graph Neural Networks; Transformers; Retail AI; Personalization; Hybrid Recommendation System

Abstract

Personalized grocery recommendation systems enhance customer experience and retail performance. While traditional methods like collaborative filtering struggle with dynamic preferences, modern deep learning and hybrid systems achieve superior accuracy and personalization. However, Traditional recommendation systems struggle with grocery-specific challenges like seasonal trends and product relationships. We propose a hybrid GNN-Transformer model that combines product graphs with purchase sequences for superior personalization and contextual understanding. The recommendation system leverages graph convolutional networks’ efficient neighborhood aggregation for local product relationships while incorporating self- attention mechanisms to capture global shopping patterns and preferences. This hybrid approach creates a powerful yet efficient recommendation system that overcomes the limitations of standalone collaborative filtering or content-based methods. The system is trained and evaluated on a large-scale grocery transaction dataset, achieving state-of- the-art recommendation accuracy while demonstrating robust performance across diverse customer segments. Experimental results show our hybrid system outperforms existing recommendation methods in accuracy, personalization, and computational efficiency. By combining graph networks and transformers, the solution delivers precise grocery suggestions while meeting real-time retail requirements. Future work will focus on optimizing the model for mobile deployment and expanding its capabilities to include meal planning recommendations.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Biswajeet Samantray
AU  - Shalini
AU  - Golda Dilip
PY  - 2025
DA  - 2025/10/31
TI  - Grocery Recommendations Using Graph Neural Networks and Transformer Models
BT  - Proceedings of the International Conference on Intelligent Systems and Digital Transformation (ICISD 2025)
PB  - Atlantis Press
SP  - 1034
EP  - 1049
SN  - 2589-4919
UR  - https://doi.org/10.2991/978-94-6463-866-0_83
DO  - 10.2991/978-94-6463-866-0_83
ID  - Samantray2025
ER  -