Abstract
The processing of Chinese names is important to the approach of Chinese word segmentation and automatic abstraction. In this paper we put forward an inverse name frequency model. Based on this model, context pattern, adjacent chain, special name table and position dependent information, we designed an effective system for automatically identifying Chinese names in texts. This paper describes the algorithm of this system, and the experiment result shows its upper recall and precision rate. Its recall rate reaches 93.75% and precision rate reaches 83.95%.
Original language | English (US) |
---|---|
Pages (from-to) | 2219-2225 |
Number of pages | 7 |
Journal | Proceedings of the IEEE International Conference on Systems, Man and Cybernetics |
Volume | 4 |
State | Published - 2001 |
Externally published | Yes |
Event | 2001 IEEE International Conference on Systems, Man and Cybernetics - Tucson, AZ, United States Duration: Oct 7 2001 → Oct 10 2001 |
Keywords
- Adjacent chain
- Context pattern
- Data sparsity
- Inverse name frequency model
- Special surname
ASJC Scopus subject areas
- Control and Systems Engineering
- Hardware and Architecture