| 一种模式匹配和统计学习相结合的文本情感分类方法 |
| |
| 引用本文： | 潘正高,张磊.一种模式匹配和统计学习相结合的文本情感分类方法[J].宿州学院学报,2013,28(1):66-68,81. |
| |
| 作者姓名： | 潘正高 张磊 |
| |
| 作者单位： | 1. 宿州学院信息工程学院,安徽宿州,234000;宿州学院智能信息处理重点实验室,安徽宿州,234000
2. 宿州市第二中学,安徽宿州,234000 |
| |
| 基金项目： | 安徽省高校省级自然科学研究一般项目"基于语义深度分析的网络舆情监测系统"，宿州学院科研开放平台项目"基于新闻实体要素的Web新闻文本分类研究" |
| |
| 摘 要： | 摘要：为了提高Web评论文本情感分类的性能，通过分析评论文本中词语之间的搭配关系，提出了一种基于模式匹配和统计学习相结合的文本情感分类方法，即将N—gram方法与情感词语搭配模式相结合，从评论文本提取特征，再利用机器学＊-2方法实现文本情感分类。实验结果表明，与传统机器学习方法相比，该方法能显著提高文本情感分类的性能。
|
| 关 键 词： | 模式匹配 特征选择 统计学习 情感倾向 |
The Method of Text Sentiment Classification Based on Pattern Matching and Statistical Learning |
| |
| PAN Zheng-gao
,
ZHANG Lei.The Method of Text Sentiment Classification Based on Pattern Matching and Statistical Learning[J].Journal of Shuzhou College,2013,28(1):66-68,81. |
| |
| Authors: | PAN Zheng-gao ZHANG Lei |
| |
| Institution: | 1.School of Information Engineering,Suzhou University,Suzhou Anhui,234000;2.Intelligent Information Processing Key Laboratory,Suzhou University,Suzhou,Anhui 234000 |
| |
| Abstract: | In order to improving the performance of Web review texts＇ sentiment classification,by annlyzing the match relationship among words of review texts,a method based on pattern matching and statistics learning is proposed. In this method,the method of N-gram and match patterns of sentiment words is combined,select fea- tures from review texts, and classify the text＇s sentiment by using a machine learning method. Experimental re- sults show that compared with traditional machine learning methods ,this method can effectively improve perfor- mance of text sentiment classification. |
| |
| Keywords: | pattern matching feature selection statistical learning sentiment tendency |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
