RaMen: Multi-Strategy Multi-Modal Learning for Bundle Construction

TL;DR

RaMen introduces a multi-strategy, multi-modal learning framework for bundle construction that captures complex relations and latent intents, outperforming existing methods by integrating explicit and implicit strategies.

Contribution

The paper presents RaMen, a novel multi-strategy approach combining explicit and implicit learning for more effective bundle construction in recommendation systems.

Findings

RaMen outperforms state-of-the-art models across multiple domains.

Explicit and implicit strategies complement each other for better bundle representations.

The multi-strategy alignment module enhances knowledge transfer and discrimination.

Abstract

Existing studies on bundle construction have relied merely on user feedback via bipartite graphs or enhanced item representations using semantic information. These approaches fail to capture elaborate relations hidden in real-world bundle structures, resulting in suboptimal bundle representations. To overcome this limitation, we propose RaMen, a novel method that provides a holistic multi-strategy approach for bundle construction. RaMen utilizes both intrinsic (characteristics) and extrinsic (collaborative signals) information to model bundle structures through Explicit Strategy-aware Learning (ESL) and Implicit Strategy-aware Learning (ISL). ESL employs task-specific attention mechanisms to encode multi-modal data and direct collaborative relations between items, thereby explicitly capturing essential bundle features. Moreover, ISL computes hyperedge dependencies and hypergraph message passing to uncover shared latent intents among groups of items. Integrating diverse strategies enables RaMen to learn more comprehensive and robust bundle representations. Meanwhile, Multi-strategy Alignment & Discrimination module is employed to facilitate knowledge transfer between learning strategies and ensure discrimination between items/bundles. Extensive experiments demonstrate the effectiveness of RaMen over state-of-the-art models on various domains, justifying valuable insights into complex item set problems.