Electric power communication network knowledge graph construction method based on BERT model

Abstract

The invention relates to an electric power communication network knowledge graph construction method based on a BERT model, and the method comprises the following steps: S1, original document library construction: selecting data of different sources and structures as a basis, and constructing the original document library; S2, label migration: labeling the original knowledge data of the power communication network, and carrying out label migration on the basis of labeling information; S3, semantic feature extraction: performing training based on a BERT model to realize semantic feature extraction of the structured document data of the power communication network; and S4, knowledge graph construction: performing named entity identification on the extracted semantic features based on a BiLSTM-CRF model, extracting knowledge concepts and interrelations, and forming the knowledge graph in cooperation with knowledge modeling. The method effectively solves the typical problems of many professional terms, few training corpora and the like, improves the accuracy of key steps such as named entity recognition and the like, and finally improves the performance of the knowledge graph.

Description

technical field [0001] The invention relates to the technical fields of informatization, automation, and intelligent maintenance of power communication networks, and in particular to a method for constructing a power communication network knowledge graph based on BERT and a BiLSTM-CRF entity recognition model. Background technique [0002] An important subsidiary facility of the modern power network is its supporting power communication network. Compared with the larger general-purpose Internet, the power communication network still maintains a relatively limited scale, and the complexity of load services is also relatively specific. However, as more and more new technologies such as the Internet of Things and artificial intelligence are introduced into the power grid construction, various devices transmit signals into the network to meet business needs, and the structure of the power communication network has become more complex. Different from the operation data of the hu...

AI Technical Summary

Problems solved by technology

Traditional entity recognition methods such as Word2Vec still focus on feature extraction of individual words rather than contextual semantic information, which leads to limitations in distinguishing specific concepts and expressing the relationship between them for further applications

Images