《计算机应用研究》|Application Research of Computers

基于互信息和邻接熵的新词发现算法

New word discovery algorithm based on mutual information and branch entropy

免费全文下载 (已被下载 次)  
获取PDF全文
作者 刘伟童,刘培玉,刘文锋,李娜娜
机构 1.山东师范大学 信息科学与工程学院,济南 250358;2.山东省分布式计算机软件新技术重点实验室,济南 250358;3.菏泽学院 计算机学院,山东 菏泽 274015
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)05-003-1293-04
DOI 10.19734/j.issn.1001-3695.2017.11.0745
摘要 如何快速高效地识别新词是自然语言处理中一项非常重要的任务。针对当前新词发现存在的问题,提出了一种从左至右逐字在未切词的微博语料中发现新词的算法。通过计算候选词语与其右邻接字的互信息来逐字扩展,得到候选新词;并通过计算邻接熵、删除候选新词的首尾停用词和过滤旧词语等方法来过滤候选新词,最终得到新词集。解决了因切词错误导致部分新词无法识别以及通过n-gram方法导致大量重复词串和垃圾词串识别为新词的问题。最后通过实验验证了该算法的有效性。
关键词 新词发现; 互信息; 邻接熵; 微博语料
基金项目 国家自然科学基金资助项目(61373148,61502151)
山东省社科规划项目(17CHLJ18,17CHLJ33,17CHLJ30)
山东省自然科学基金资助项目(ZR2014FL010)
山东省教育厅基金资助项目(J15LN34)
本文URL http://www.arocmag.com/article/01-2019-05-003.html
英文标题 New word discovery algorithm based on mutual information and branch entropy
作者英文名 Liu Weitong, Liu Peiyu, Liu Wenfeng, Li Nana
机构英文名 1.School of Information Science & Engineering,Shandong Normal University,Jinan 250358,China;2.Shandong Provincial Key Laboratory for Distributed Computer Software Novel Technology,Jinan 250358,China;3.School of Computer Science,Heze University,Heze Shandong 274015,China
英文摘要 How to identify new words quickly and efficiently is a very important task in natural language processing. Aiming at the problems existing in the discovery of new words, this paper proposed an algorithm for word-finding new words verbatim from left to right in the uncut word Weibo corpus. One way to get a candidate new word was by computing the candidate word and its right adjacent word mutual information to expand word by word; there were some ways to filter candidate new words to get new word sets. It included methods included calculating the branch entropy, deleting stop words contained in the first or last word of each candidate new word and deleting old words included in the candidate new word set. It solved the problem that some new words could not be recognized due to the mistakes in the word segmentation and it also solved the problem that the large number of repetitive word strings and rubbish words strings generated by the n-gram method were identified as new words. Finally, experiments verify the effectiveness of the algorithm.
英文关键词 new word discovery; mutual information; branch entropy; microblog corpus
参考文献 查看稿件参考文献
 
收稿日期 2017/11/20
修回日期 2018/1/10
页码 1293-1296
中图分类号 TP301.6
文献标志码 A