基于规则的藏文音节纠正算法  

Rules Based Tibetan Syllable Correction

作　　者：普尺 吕慧 李妍[1] 多拉 周庆国[1] Puchi;LYU Hui;LI Yan;Dolha;ZHOU Qingguo(College of Information Science and Engineering,Lanzhou University,Lanzhou,Gansu 730000,China;College of Electrical Engineering,Northwest Minzu University,Lanzhou,Gansu 730000,China;State Key Laboratory of Tibetan Intelligent Information Processing and Application,Qinghai Normal University,Xining,Qinghai 810000,China)

机构地区：[1]兰州大学信息科学与工程学院,甘肃兰州730000 [2]西北民族大学电气工程学院,甘肃兰州730000 [3]青海师范大学省部共建藏语智能信息处理及应用国家重点实验室,青海西宁810000

出　　处：《中文信息学报》2025年第1期28-36,共9页Journal of Chinese Information Processing

基　　金：国家重点研究与发展计划(2020YFC0832500);国家自然科学基金(61402210,62266037)。

摘　　要：藏文语料库的质量是影响藏文信息处理的重要因素。通过分析藏文音节拼写规则,该文构建了针对藏文规则音节和非规则音节拼写检查的音节规则库和非规则音节字典(包括未登录词音译、本体藏文和梵音转写),并基于此提出基于规则的藏文音节纠正算法。该算法利用音节规则库和非规则音节字典对识别后的藏文音节进行拼写检查,通过规则音节构件组合规则来实现藏文音节纠正。实验结果表明,该文提出的藏文规则音节和非规则音节拼写检查算法的F_(1)值达到100%,而规则音节纠正算法的宏平均准确率达到80.43%。The quality of the Tibetan corpus significantly impacts Tibetan information processing.This paper analyzes Tibetan syllable spelling rules and establishes a syllable rule base along with an irregular syllable dictionary.The dictionary includes transliteration of unknown words,ontology Tibetan,and Sanskrit transliteration.It is designed for both regular and irregular Tibetan syllables,facilitating spelling checking.Additionally,a rule-based Tibetan syllable correction algorithm is proposed.This algorithm utilizes the syllable rule base and the irregular syllable dictionary to check recognized Tibetan syllables.It corrects regular syllables effectively by employing combination rules of regular syllable components.Experimental results demonstrate that the proposed method achieves a 100%F_(1) value,with the macro-average accuracy of the regular syllable correction algorithm reaching 80.43%.

关 键 词：语料库 藏文音节 拼写检查 音节纠正 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]