Improving Sequential Recommendation with Attribute-augmented Graph Neural Networks

TL;DR

This paper introduces Murzim, a graph neural network model that leverages item attribute information to improve sequential recommendations, demonstrating superior performance and real-world deployment in a large streaming platform.

Contribution

The paper presents Murzim, a novel attribute-augmented GNN model that effectively incorporates item attributes into sequential recommendation systems.

Findings

Murzim outperforms state-of-the-art methods in recall and MRR.

Murzim successfully utilizes item attribute information for better recommendations.

Murzim is deployed in a large-scale streaming platform, recommending videos to tens of thousands of users.

Abstract

Many practical recommender systems provide item recommendation for different users only via mining user-item interactions but totally ignoring the rich attribute information of items that users interact with. In this paper, we propose an attribute-augmented graph neural network model named Murzim. Murzim takes as input the graphs constructed from the user-item interaction sequences and corresponding item attribute sequences. By combining the GNNs with node aggregation and an attention network, Murzim can capture user preference patterns, generate embeddings for user-item interaction sequences, and then generate recommendations through next-item prediction. We conduct extensive experiments on multiple datasets. Experimental results show that Murzim outperforms several state-of-the-art methods in terms of recall and MRR, which illustrates that Murzim can make use of item attribute information to produce better recommendations. At present, Murzim has been deployed in MX Player, one of India's largest streaming platforms, and is recommending videos for tens of thousands of users.