基于最大-最小相似度学习方法的文本提取 - citeseerx.pdf

ISSN 1000-9825, CODEN RUXUEW E-mail: jos@ Journal of Software , Vol.19, No.3, March 2008, pp.621−629 DOI: 10.3724/SP.J.1001.2008.00621 Tel/Fax: +86-10 © 2008 by Journal of Software . All rights reserved. ∗ 基于最大-最小相似度学习方法的文本提取 1,2 1,3+ 1 付 慧 , 刘峡壁 , 贾云得 1(北京理工大学 计算机科学与技术学院 智能信息技术北京市重点实验室,北京 100081) 2(北京林业大学 信息学院,北京 100083) 3( 中国科学院 自动化研究所 模式识别国家重点实验室,北京 100080) Text Extraction Based on Maximum-Minimum Similarity Training Method FU Hui1,2, LIU Xia-Bi1,3+, JIA Yun-De1 1(Beijing Laboratory of Intelligent Information Technology, School of Computer Science and Technology, Beijing Institute of Technology, Beijing 100081, China) 2(School of Information Technology, Beijing Forestry University, Beijing 100083, China) 3(National Laboratory of Pattern Recognition, Institute of Automation, The Chinese Academy of Sciences, Beijing 100080, China) + Corresponding author: Phn: +86-10 Fax: +86-10 E-mail: liuxiabi@, Fu H, Liu XB, Jia YD. Text extraction based on maximum-minimum similarity training method. Journal of Software, 2008,19(3):621−629. /1000-9825/19/621.htm Abstract : This paper proposes a maximum-minimum similarity training algorithm to optimize the parameters in the effective method of text extraction based on Gaussian mixture modeling of neighbor characters. The maximum-minimum similarity training (MMS) methods optimize recognizer performance through maximizing the similarities of positive samples and minimizing the similarities of negative samples. Based on this approach to discriminative training, it def