摘要： 研究名词与量词的关系对于自然语言处理有很大的帮助，目前用于自然语言处理用的量词名词词典是由人工方法实现的，并由人工进行维护，有更新慢、收藏词少等缺点。该文利用数据挖掘技术对大规模真实文本进行分析，挖掘量词与名词珠深层关系，有效地实现了对人工量词名词词典的扩充。

关键词: 数据挖掘, 互信息, 关联规则挖掘, 量词名词词典

Abstract: A Quantifier-nonu dictionary plays an important role in many natural language processing applications,such as part-of-speech tagging,syntax analysis and machine translation. Currently, the quantifier-nonu dictionary is constructed by linguists.It can not be updated frequently and many lexicon entries may not be available. In this paper, a data-mining approach based on large-scale corpus analysis is used to extend a quantifier-noun dictionary. The main purpose is to make a quantifier-nonu dictionary more applicable.Experimental results show the approach is effective and suitable.

中图分类号: 

• TP391.1