Year: 2023
Type: conference-paper
Abstract: In order to extract named entities from power equipment text, an entity recognition model based on character pair link is proposed in this paper. This model decodes links based on characters and incorporates lexical information to improve the model prediction effect. In order to extract entities from the text of power equipment, the original corpus is constructed by power equipment-related texts and actual fault cases, and the entity recognition data set of power equipment is constructed based on entity annotation. Finally, the entity recognition model is used to conduct experiments on the data set. The experimental results show that the proposed model improves the entity recognition rate by 29.2% and 13.9%, respectively, compared with the existing algorithms, and thus improves the construction efficiency of knowledge graph.
Authors: Jiangbin Yu, Jun Tao, Qing Guo, Meng Xue, Wenping Nie, Xudong Huang
Institution: Anhui Jiyuan Software
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4396648484