摘要
提出了一种汉语文本切分和词性标注相融合的一体化分析的统计模型,并应用动态规划算法与A*解码算法相结合的二次搜索算法,实现了一个基于该模型的汉语词法分析器。初步的开放测试表明,该分析器的分词准确率和词性标注正确率分别可达98.67%和95.49%。
In this paper,we present a stochastic model integrating Chinese word segmentation with part-of-speech tagging.We also develop a Chinese lexical analyzer using a two-way searching algorithm which incorporates backward dynamic programming algorithm into A*decode algorithm.The primary experiment proved that the overall accuracy of the proposed analyzer is 98.67% for segmentation and 95.49% for POS tagging respectively.
出处
《计算机应用研究》
CSCD
北大核心
2001年第7期24-26,共3页
Application Research of Computers
基金
国家"863"计划资助项目(863-ZT-03-02-3)