Leveraging Multimodal Features and Item-level User Feedback for Bundle Construction

TL;DR

This paper introduces CLHE, a contrastive learning-based hierarchical encoder, to improve bundle construction by effectively integrating multimodal features and user feedback, addressing data sparsity and cold-start issues.

Contribution

The paper proposes a novel CLHE framework that unifies multimodal and multi-item features using self-attention and contrastive learning for better bundle construction.

Findings

Outperforms state-of-the-art methods on four datasets

Effectively handles modality missing and noise issues

Enhances representation learning for sparse and cold-start data

Abstract

Automatic bundle construction is a crucial prerequisite step in various bundle-aware online services. Previous approaches are mostly designed to model the bundling strategy of existing bundles. However, it is hard to acquire large-scale well-curated bundle dataset, especially for those platforms that have not offered bundle services before. Even for platforms with mature bundle services, there are still many items that are included in few or even zero bundles, which give rise to sparsity and cold-start challenges in the bundle construction models. To tackle these issues, we target at leveraging multimodal features, item-level user feedback signals, and the bundle composition information, to achieve a comprehensive formulation of bundle construction. Nevertheless, such formulation poses two new technical challenges: 1) how to learn effective representations by optimally unifying multiple features, and 2) how to address the problems of modality missing, noise, and sparsity problems induced by the incomplete query bundles. In this work, to address these technical challenges, we propose a Contrastive Learning-enhanced Hierarchical Encoder method (CLHE). Specifically, we use self-attention modules to combine the multimodal and multi-item features, and then leverage both item- and bundle-level contrastive learning to enhance the representation learning, thus to counter the modality missing, noise, and sparsity problems. Extensive experiments on four datasets in two application domains demonstrate that our method outperforms a list of SOTA methods. The code and dataset are available at https://github.com/Xiaohao-Liu/CLHE.