GRAINRec: Graph and Attention Integrated Approach for Real-Time Session-Based Item Recommendations

TL;DR

GRAINRec is a real-time session-based recommendation model that combines graph and attention mechanisms, improving recommendation relevance and scalability in online retail settings.

Contribution

It introduces a novel integrated graph and attention model for real-time session recommendations with a heuristic for scalable inference.

Findings

1.5% average improvement in offline metrics

10% increase in click-through rate in A/B tests

9% increase in attributable demand

Abstract

Recent advancements in session-based recommendation models using deep learning techniques have demonstrated significant performance improvements. While they can enhance model sophistication and improve the relevance of recommendations, they also make it challenging to implement a scalable real-time solution. To addressing this challenge, we propose GRAINRec: a Graph and Attention Integrated session-based recommendation model that generates recommendations in real-time. Our scope of work is item recommendations in online retail where a session is defined as an ordered sequence of digital guest actions, such as page views or adds to cart. The proposed model generates recommendations by considering the importance of all items in the session together, letting us predict relevant recommendations dynamically as the session evolves. We also propose a heuristic approach to implement real-time inferencing that meets Target platform's service level agreement (SLA). The proposed architecture lets us predict relevant recommendations dynamically as the session evolves, rather than relying on pre-computed recommendations for each item. Evaluation results of the proposed model show an average improvement of 1.5% across all offline evaluation metrics. A/B tests done over a 2 week duration showed an increase of 10% in click through rate and 9% increase in attributable demand. Extensive ablation studies are also done to understand our model performance for different parameters.