Sentiment polarity analysis on the microblogging hot topic can better understand the Internet user's attitude and tendency toward a specific event, so that government can effectively guide the public opinion. Different from other methods, the sentiment polarity analysis method put forward in this paper gives full consideration to the expression characteristics of microblogging, a particular network media. It particularly considered the network languages and emoticons when calculated the value of sentiment polarity. In this method, we calculate the value of sentiment polarity of each word firstly by combining Pointwise Mutual Information (PMI) and HowNet. Secondly, modify the sentiment polarity value of a word through syntactic dependencies. Finally, accumulate the sentiment polarity value of each word, so the sentiment polarity value of a microblogging can be obtained. The sentiment polarity value of a hot topic can be obtained in the end, through accumulating the sentiment polarity value for all microblogging in a hot topic. Contrast experiment results show that, the method can obtain the sentiment polarity value of a hot topic more accurately and effectively.
[1]
Huang Tin.
Emotional Orientation Analysis of Microblog Based on Graph Model and Multiple Classifiers
,
2015
.
[2]
Yang Shen,et al.
Emotion mining research on micro-blog
,
2009,
2009 1st IEEE Symposium on Web Society.
[3]
Jian-li Ding,et al.
Orientation analysis of Web reviews: Orientation analysis of Web reviews
,
2010
.
[4]
Michael L. Littman,et al.
Measuring praise and criticism: Inference of semantic orientation from association
,
2003,
TOIS.
[5]
Dai Lin Shi Han-xiao Huang Qi Li Xiao-jun.
Survey on sentiment orientation analysis of texts
,
2011
.
[6]
Yang Zhihao.
Text Orientation Identification Based on Semantic Comprehension
,
2007
.
[7]
He Feng-ying.
Orientation analysis for Chinese blog text based on semantic comprehension
,
2011
.
[8]
Liu Lu,et al.
Empirical study of sentiment classification for Chinese microblog based on machine learning
,
2012
.