重庆理工大学学报（自然科学版）2013，Vol.27Issue(1)：51-56,6.DOI:10.3969/j.issn.1674-8425(z).2013.01.011
中文微博情感分析中主客观句分类方法
A Study on the Classification Approach for Chinese MicroBlog Subjective and Objective Sentences
杨武 1宋静静 1唐继强1
作者信息
- 1. 重庆理工大学计算机科学与工程学院,重庆400054
- 折叠
摘要
Abstract
As a result of the rapidly increasing number of microblog users, sentiment analysis on mi-croblog data has attracted more and more attention. The first step of sentiment analysis is to effectively identify the subjective information of the texts. How to effectively extract subjective information and analyze opinions on microblog has become a new popular research topic. In this paper, we use Naive Bayes classifier to give a solution for the classification between subjective and objective sentence in Chinese microblog. First, we present the difference between microblog and other texts, and then we select some subjective and objective lexical clues for Chinese microblog. Second, we introduce 2-POS model and its feature selection methods. Third, we take key words and subjective lexical clues as semantic features, 2-POS as grammatical features, and then use them to construct Naive Bayes classifier respectively. The experiment results show that the method that uses both the semantic features and the grammatical features achieves the best performance.关键词
微博/主客观分类/主客观线索/2-POS模式/朴素贝叶斯Key words
microblog/ subjective and objective classification/ subjective and objective clues/ 2-POS features/ naive Bayes分类
信息技术与安全科学引用本文复制引用
杨武,宋静静,唐继强..中文微博情感分析中主客观句分类方法[J].重庆理工大学学报（自然科学版）,2013,27(1):51-56,6.
