一种融合自适应增强语义信息的知识图谱补全方法

A Knowledge Graph Completion Method with Adaptive Enhancement of Semantic Information

翻译模型在进行知识图谱补全的过程中往往会忽略三元组中的语义信息。为弥补这一缺陷,本文构造了一种融合自适应增强语义信息的知识图谱补全方法。通过微调BERT模型获取三元组中的语义信息,并对高纬度向量做降维处理,最后运用注意力机制生成语义信息软约束规则,将该规则添加至原翻译模型中实现语义信息的自适应增强。经实验对比,本文所提方法较原翻译模型在数值上约提升2.6%,验证了方法的合理性与有效性。

Translation models tend to ignore the semantic information in triads in the process of knowledge graph complementation.To remedy this shortcoming,this paper constructs a knowledge graph complementation method that incorporates adaptively enhanced semantic information.The semantic information in the triad is obtained by fine-tuning the BERT model,and the high-latitude vectors are dimensionally reduced.Finally,the attention mechanism is applied to generate soft constraint rules for semantic information,and the rules are added to the original translation model to achieve adaptive enhancement of semantic information.After the experimental comparison,the proposed method improves about 2.6%in value compared with the original translation model,which verifies the reasonableness and effectiveness of the method.