Simulated annealing clustering of Chinese words for contextual text recognition

被引:12
作者
Chang, CH
机构
[1] E000/CCL, Building 11, Indust. Technol. Research Institute, Chutung
关键词
simulated annealing; word clustering; perplexity; optical character recognition; contextual postprocessing;
D O I
10.1016/0167-8655(95)00080-1
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Simulated annealing clustering algorithms are applied in discovering word classes in a Chinese class n-gram model, which can be used for contextual postprocessing of handwritten Chinese character recognition. Experimental results show that the proposed model achieves much better performance than the dictionary-based models and outperforms the well-known inter-word character bigram model while using less storage.
引用
收藏
页码:57 / 66
页数:10
相关论文
共 17 条
[1]  
Brown P. F., 1992, Computational Linguistics, V18, P467
[2]  
CHANG CH, 1993, P ROCLING CHITOU, V6, P57
[3]  
CHANG CH, 1993, P NATURAL LANGUAGE P, P319
[4]  
CHOU BH, 1992, P ROCLING, V5, P261
[5]  
Chou S.-L., 1993, Proceedings of the Second International Conference on Document Analysis and Recognition (Cat. No.93TH0578-5), P474, DOI 10.1109/ICDAR.1993.395692
[6]  
JARDINO M, 1993, P ICASSP 93, V2, P41
[7]   OPTIMIZATION BY SIMULATED ANNEALING [J].
KIRKPATRICK, S ;
GELATT, CD ;
VECCHI, MP .
SCIENCE, 1983, 220 (4598) :671-680
[8]   EXPERIMENTS IN PROJECTION AND CLUSTERING BY SIMULATED ANNEALING [J].
KLEIN, RW ;
DUBES, RC .
PATTERN RECOGNITION, 1989, 22 (02) :213-220
[9]  
Lee H.-J., 1993, Proceedings of the Second International Conference on Document Analysis and Recognition (Cat. No.93TH0578-5), P72, DOI 10.1109/ICDAR.1993.395779
[10]  
Lee Lin-Shan, 1993, P IEEE INT C AC SPEE, V2, P503