关键词: 知识图谱补全, 多模态融合, 本体信息, 结构信息, 决策学习

Abstract: Knowledge graph completion aims to expand and enhance knowledge graphs by predicting missing triples. Multi-modal knowledge graph completion integrates entity ontology information such as entity descriptions, entity images, and entity attributes to obtain more accurate entity representations. Existing research projects different modalities into a unified space to obtain joint representations of entities, then combine knowledge graph structural information for predictions. However, existing methods have difficulty capturing the complex interactions between entity background knowledge when fusing multi-modal information, which inevitably leads to information loss and insufficient feature extraction capabilities; overfitting and limited entity relation interactions restrict the performance of 2D convolution models, making it difficult to integrate knowledge graph structural information. Therefore, this paper used a multi-level fusion knowledge graph completion model. To fully integrate entity multimodal information, three different fusion methods are simultaneously used, along with joint decision learning, aiming to combine the complementary information provided by different multi-modal fusion methods to obtain rich and diverse entity representations. To fully integrate knowledge graph structural information, feature generalization is proposed to alleviate the overfitting issues of 2D convolution models, combined with feature reshaping to enhance interactions between entities and relations. Experiments on multiple public datasets demonstrate the superior performance of the proposed method.

Key words: knowledge graph completion, multi-model fusion, ontology information, structural information, decision learning