 |
|
|
|
 |
 |
 |
|
 |
|
 |
|
|
王挺,陈火旺,杨谊,史晓东.一种自适应词性标注方法.软件学报,1997,8(12):937-943 |
一种自适应词性标注方法 |
AN ADAPTIVE METHOD FOR PART-OF-SPEECH TAGGING |
修订日期:1997-01-13 |
DOI: |
中文关键词: 语料 词性标注 隐马尔可夫模型 |
英文关键词:Corpus part-of-speech tagging hidden Markov model |
基金项目:本文研究得到国家863高科技项目和校预研项目基金资助. |
|
摘要点击次数: 3548 |
全文下载次数: 3562 |
中文摘要: |
本文针对词性标注的问题,修改了经典的隐马尔可夫模型HMM(hiddenMarkovmodel)参数估算方法,使得模型参数能够随着新语料的增加而动态地进行调整.从已标注或未标注的语料中获取知识以提高模型的准确性. |
英文摘要: |
In this paper, the authors modified the classical HMM (hidden Markovmodel) method so as to train the model whenever the new corpus is introduced and no matter the corpus is tagged or not. Therefore, the information can be learned from the corpusto improve the accuracy of the model. |
HTML 下载PDF全文 查看/发表评论 下载PDF阅读器 |
|
|
|
|
|
|
 |
|
|
|
|
 |
|
 |
|
 |
|