基于自然语言处理模型的多音字对汉语拼音字母排序的影响研究  

Study on the effect of polyphone in Pinyin alphabetic sorting based on natural language processing model

作　　者：王云凯[1] 王萍[1] 

机构地区：[1]西南财经大学经济信息工程学院,成都611130

出　　处：《西南民族大学学报（自然科学版）》2012年第3期479-482,共4页Journal of Southwest Minzu University(Natural Science Edition)

摘　　要：在对汉字进行拼音字母统计分析的时候经常会遇到对多音字的处理问题,到底多音字对汉字拼音字母统计结果的影响有多大呢?选取总字数达1200多万的各类网络文献为语料,采用多音字取常用读音和多音字细化处理两种统计方式对语料的汉语拼音字母进行统计,通过对两种统计结果进行对比分析,得出多音字对汉字拼音字母统计影响很小这一结论.In analysis and statistics processing of Pinyin alphabetic, usually, there is a confusion that how much influence of polyphone have on the result. So it is necessary to get the degree of influence of polyphone to the result of Pinyin alphabetic statistics. This paper chooses various types of network literature amounting to 12 million words as corpus, selects common pronunciation way and pronunciation refinement way to measure every character＇s rate in corpus respectively. Through comparing and analyzing the two results, it is concluded that the effect of polyphone in Pinyin alphabetic sorting is slim.

关 键 词：自然语言处理 自然语言统计 汉字拼音 多音字 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]