An Electrical Power Text Entity Extraction Method Based on BERT Model

Yaqin Luo, Yan Huang, Bailing Zhou, Xiaoli Liu, Yuyang Tang, Yang Liu

Abstract


Most important electrical power information on the Internet is hidden in natural language text and is difficult to be directly processed by the computer. In this paper, an electrical power text entity extraction method is proposed to extract the key electrical power information that people need from a large number of natural language texts. The extraction of power entities provides a large amount of data for the establishment of the power knowledge graph, which facilitates the daily life and industry production. In this paper, the method is tested and evaluated on the text data of power science and technology achievements, which proves that electrical power text entity extraction method has high accuracy.


DOI
10.12783/dtmse/ameme2020/35602

Refbacks

  • There are currently no refbacks.