location: Current position: Home >> Scientific Research >> Paper Publications

基于半监督隐马尔科夫模型的汉语词性标注研究

Hits:

Indexed by:期刊论文

Date of Publication:2015-12-15

Journal:小型微型计算机系统

Included Journals:PKU、ISTIC、CSCD

Volume:36

Issue:12

Page Number:2813-2816

ISSN No.:1000-1220

Key Words:词性标注;词向量;词语相似度;迭代训练

Abstract:提出一种基于词语相似度计算的半监督隐马尔科夫词性标注方法.首先,利用小规模的训练语料进行半监督隐马尔科夫学习,通过反复迭代不断扩充语料,增强隐马尔科夫的标注效果;然后,通过计算词语相似度的方法,给测试语料中每个未登录词都标上候选词性;最后,在隐马尔科夫标注时,不是选取一条最佳路径,而是选取两条最佳路径,通过二次选择,以此得到标注结果.实验结果证明,该方法与传统的隐马尔科夫标注方法相比提高了约2.60%,汉语词性标注准确率达到了95.65%.

Pre One:Biomedical named entity recognition based on recurrent neural networks with different extended methods

Next One:Combining Feature-Based and Instance-Based Transfer Learning Approaches for Cross-Domain Hedge Detection with Multiple Sources