Research on the Construction of Heterogeneous Resource Data Relation Extraction System Based on Computer Knowledge Graph
Fan Zhang, Mao Hangyin, Cao Ying, Wei Li, Nie Qing · 2023
In this paper, multi-source heterogeneous online review data is used to construct a knowledge graph framework. The pattern layer constructs the conceptual framework of knowledge map by centering on the information source, content and form of online reviews. word2vec is used to obtain entities, relationships and attributes from heterogeneous texts, and data fusion and knowledge graph analysis are carried out. A two-stage knowledge graph model construction method is proposed to solve the multi-chain knowledge graph model of service value chain. The system realizes the storage of knowledge graph. Finally, the effectiveness of the knowledge map for online review research and mining is verified systematically. The results reveal the characteristics of multi-source heterogeneous online review data.