Fusing visual and textual content for knowledge graph embedding via dual-track model

Yancong Li, Xiaoming Zhang, Fang Wang, Bo Zhang, Feiran Huang

Abstract: Highlights•A novel multi-modal fusion model for knowledge graph embedding is proposed.•The model can fuses multi-modal information to learn more effective embeddings.•The model can be used to fuse more types of data with specified decoders.•Experiments show the superiority of the model in performance and flexibility.