Recognition of the Name of Quality Inspection Institutions Based on CRF and Rules. (July 2020)

• Record Type: Journal Article
• Authors: Ning, Xiuli; Su, Xueyan; Peng, Bin
• Is Part Of: Journal of physics. Volume 1584(2020)
• Publication Date: 2020-07
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1584/1/012063
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: In recent years, the case of injuries, deaths and property losses caused by product quality and safety issues occurred frequently. The analysis and processing of product quality and safety data can provide data support for government departments in decision-making, which is significant for improving product quality and safety in general. Institution name recognition is an important link in the analysis and processing of quality inspection data. A recognition method which combined CRF and rules for the name of quality inspection institution is put forward in this paper. Firstly, summarizing the rules and features of the name of quality inspection institution by the analysis of the composition characteristics, and then the entity recognition model based on statistics and rules is constructed in combination with CRF. The experimental results show that the method proposed this paper has high accuracy, recall rate and F value for the quality inspection institution names of different scales.