KNOWLEDGE FUSION METHOD AND APPARATUS BASED ON DATA RELATIONSHIP ANALYSIS, AND COMPUTER DEVICE AND STORAGE MEDIUM

摘要

The present application relates to the technical field of knowledge graphs, in particular to a knowledge fusion method and apparatus based on data relationship analysis, and a computer device and a storage medium. The method comprises: acquiring data to be analyzed, extracting, according to the data structure of the data to be analyzed, entity data in the data to be analyzed, and performing entity alignment on the entity data to obtain preliminarily fused data; acquiring a relationship path in the preliminarily fused data, and constructing, according to the relationship path, a feature vector of relationship classification; determining, according to the feature vector, relational data between unconnected entity data in the preliminarily fused data to obtain fused data to be detected; and performing problem information query on the fused data to be detected, and removing problem information in order to obtain final fused data. In the present application, by means of adding a step of knowledge error correction to a knowledge fusion process, erroneous data generated in the knowledge fusion process can be effectively eliminated, thereby improving the efficiency when performing knowledge extraction on the knowledge graph formed after knowledge fusion.