Title :
A Chinese name identifying system based on inverse name frequency model and rules
Author :
Ji, Heng ; Luo, Zhensheng
Author_Institution :
Lab. of Computational Linguistics, Tsinghua Univ., Beijing, China
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%
Keywords :
character recognition; natural languages; text analysis; Chinese names; Chinese word segmentation; adjacent chain; automatic abstraction; context pattern; data sparsity; inverse name frequency model; position dependent information; special name table; Computational linguistics; Context modeling; Error analysis; Frequency; Inverse problems; Statistics;
Conference_Titel :
Systems, Man, and Cybernetics, 2001 IEEE International Conference on
Conference_Location :
Tucson, AZ
Print_ISBN :
0-7803-7087-2
DOI :
10.1109/ICSMC.2001.972886
