计算机技术与发展2017，Vol.27Issue(8)：43-46,51,5.DOI:10.3969/j.issn.1673-629X.2017.08.009
一种基于词树的高效解码算法
An Efficient Decoding Algorithm Based on Word Tree
摘要
Abstract
Phonetic conversion is an important aspect of Chinese language information processing,which has been widely used in speech recognition,Chinese Pinyin input and so on.In order to find an effective syllable-to-character decoding algorithm,an efficient decoding algorithm is proposed based on the study of phonetic word segmentation,the word tree theory and the analysis of word tree solving.It uses zero probability reassessment,path pruning,processing of polyphonic words to realize the syllable-to-character conversion generally by pruning of word tree,processing of common words and processing of polyphonic words in the decoding process.In order to verify the validity and feasibility of the proposed algorithm,the contrast experiments on three sets of data provided by Xinjiang Uygur Autonomous Region Science and Technology Program,Multilingual Ethnic Cultural Information Resource Processing and Sharing Service Platform,have been conducted.The experimental results show that it has achieved 97.78% conversion accuracy,which is superior to other traditional algorithms.关键词
拼音分词/词树/语言模型/n-gram模型/音字转换Key words
phonetic word segmentation/lexicon tree/language model/n-gram model/Pinyin-Chinese character transform分类
信息技术与安全科学引用本文复制引用
张志强,张太红,董峦..一种基于词树的高效解码算法[J].计算机技术与发展,2017,27(8):43-46,51,5.基金项目
新疆维吾尔自治区科技计划项目(2015X0106) （2015X0106）
