Construction and design of knowledge graph in the power field
ZHANG Senda
CUI Xin
QU Yangang
SUN Jian
XIE Junyi
Abstract：To address the limitations of traditional methods in handling multi-source heterogeneous data for knowledge extraction in the power sector,the article constructs a power-specific knowledge graph to enhance the intelligence of dispatching.It employs a long short-term memory(LSTM)conditional random field(CRF)model for entity extraction and uses a piecewise convolutional neural network(PCNN)for relation extraction,combined with the Neo4j graph database for storing and querying the knowledge graph.The results show that the LSTM-CRF model achieves an F1 score of 76.58% for entity extraction tasks,while the PCNN model performs excellently in relation extraction tasks.
Keywords：power fieldknowledge graphLSTMCRFPCNN
Publication Date：2025-10-25
Pages：4( 66-69 )