Search papers, labs, and topics across Lattice.
The paper introduces E-MMKGR, a novel framework that constructs an e-commerce multimodal knowledge graph (E-MMKG) to address the limitations of existing multimodal recommender systems in modality extensibility and task generalization. E-MMKGR learns unified item representations via GNN-based propagation and KG-oriented optimization on the E-MMKG, enabling a shared semantic foundation for various downstream tasks. Empirical results on Amazon datasets demonstrate significant improvements in recommendation (up to 10.18% in Recall@10) and product search (up to 21.72% over vector-based retrieval).
E-commerce recommendations and product search get a serious boost from a unified multimodal knowledge graph that beats existing methods by up to 21%.
Multimodal recommender systems (MMRSs) enhance collaborative filtering by leveraging item-side modalities, but their reliance on a fixed set of modalities and task-specific objectives limits both modality extensibility and task generalization. We propose E-MMKGR, a framework that constructs an e-commerce-specific Multimodal Knowledge Graph E-MMKG and learns unified item representations through GNN-based propagation and KG-oriented optimization. These representations provide a shared semantic foundation applicable to diverse tasks. Experiments on real-world Amazon datasets show improvements of up to 10.18% in Recall@10 for recommendation and up to 21.72% over vector-based retrieval for product search, demonstrating the effectiveness and extensibility of our approach.
