CASIA OpenIR  > 模式识别国家重点实验室  > 自然语言处理
基于多样化特征的中文微博情感分类方法研究
张志琳; 宗成庆
Source Publication中文信息学报
2015-07
Volume29Issue:7Pages:134-143
AbstractMicro blog,a new information—sharing platform,is now playing an important role in people’S daily live with the rise of Web 2.0.And micro blog sentiment analysis research also attracts more attention in recent years. This paper provides an in—depth analysis on the difference of feature representation and feature selection between the traditional sentiment classification and micro blog sentiment analysis.To avoid the drawbacks of feat ure selection of existing methods,we propose three simple but effective approaches for feature representation and selection,inclu— ding the lexicalization hashtag feature,the sentiment word feature,and the probabilistic sentiment lexicon feature. Experimental results show that our proposed methods significantly boost the micro blog sentiment classification ac— curacy from 73.1 7%tO 84.1 7%,outperforming the state—of—the-art method significantly
Keyword中文微博 情感分类 机器学习 特征选择
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/19838
Collection模式识别国家重点实验室_自然语言处理
Affiliation中国科学院自动化研究所
Recommended Citation
GB/T 7714
张志琳,宗成庆. 基于多样化特征的中文微博情感分类方法研究[J]. 中文信息学报,2015,29(7):134-143.
APA 张志琳,&宗成庆.(2015).基于多样化特征的中文微博情感分类方法研究.中文信息学报,29(7),134-143.
MLA 张志琳,et al."基于多样化特征的中文微博情感分类方法研究".中文信息学报 29.7(2015):134-143.
Files in This Item: Download All
File Name/Size DocType Version Access License
张志琳基于多样化特征的中文微博情感分类方(2117KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[张志琳]'s Articles
[宗成庆]'s Articles
Baidu academic
Similar articles in Baidu academic
[张志琳]'s Articles
[宗成庆]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[张志琳]'s Articles
[宗成庆]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 张志琳基于多样化特征的中文微博情感分类方法研究.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.