Abstract
The development of mobile internet has led to a massive amount of data being generated from mobile devices daily, which has become a source for analyzing human behavior and trends in public sentiment. In this paper, we build a system called MoSa (Mobile Sentiment analysis) to analyze this data. In this system, sentiment analysis is used to analyze news comments on the THAAD (Terminal High Altitude Area Defense) event from Toutiao by employing algorithms to calculate the sentiment value of the comment. This paper is based on HowNet; after the comparison of different sentiment dictionaries, we discover that the method proposed in this paper, which use a mixed sentiment dictionary, has a higher accuracy rate in its analysis of comment sentiment tendency. We then statistically analyze the relevant attributes of the comments and their sentiment values and discover that the standard deviation of the comments’ sentiment value can quickly reflect sentiment changes among the public. Besides that, we also derive some special models from the data that can reflect some specific characteristics. We find that the intrinsic characteristics of situational awareness have implicit symmetry. By using our system, people can obtain some practical results to guide interaction design in applications including mobile Internet, social networks, and blockchain based crowdsourcing.
Subject
Physics and Astronomy (miscellaneous),General Mathematics,Chemistry (miscellaneous),Computer Science (miscellaneous)
Reference18 articles.
1. Enhancing network capacity by weakening community structure in scale-free network
2. Emotion Analysis on Text Words and Sentences based on SVM;Jing;Comput. Appl. Softw.,2011
3. Research on analyzing sentiment of texts based on k-nearest neighbor algorithm;Na;Comput. Eng. Des.,2012
4. Empirical study of sentiment classification for Chinese microblog based on machine learning;Zhiming;Comput. Eng. Appl.,2012
5. Improved information gain text feature selection algorithm based on word frequency information;Hui;Comput. Appl.,2014
Cited by
7 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献