摘要： 介绍了一个对汉语语料进行切分标注的系统——SEGPOS。该系统采用模块化设计,具有灵活的算法实现、替换机制,并能动态扩展各类数据资源。为开发切分、标注算法,进行多种算法的横向比较研究,提供了一个良好的平台。给出了在SEGPOS上实现的多种切分、标注算法的测试结果和分析。

关键词: 切分, 标注, 自然语言理解

Abstract: This paper presents a Chinese corpus segmentation & part of speech tagging system (SEGPOS). This system provides a platform to implement many kinds of segmentation and part of speech tagging algorithms. It can be used to research the difference between two algorithms. At the end, this paper gives a report about the result of using these algorithms on SEGPOS.

中图分类号: 

• TP391.1