Abstract
Analysis on microblog sentiment is very important for microblog monitoring and guidance of public opinion. According to the problem that precision of Chinese microblog sentiment analysis was low, this paper proposed a model for analysis on Chinese microblog sentiment based on syntax parsing and the support vector machine algorithm. Firstly, this paper built a fundamental dictionary of emotion based on existing emotional words resources, and expanded the dictionary by computing similarity of words. Then, it computed emotional values of microblogs by syntax parsing and judged their emotional tendencies. Finally, it selected a certain percentage of positive and negative emotional microblogs as a training set. It utilized the support vector machine algorithm to classify microblogs that didn’t belong to the training set and got all emotional tendencies of microblogs. Experimental results showed that the model proposed by this paper could achieve 77.3% precision. So the model is effective.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Tan, S., Liu, K., Wang, S., et al.: Overview of Chinese Opinion Analysis Evaluation 2013. In: Proceeding of the Fifth Chinese Opinion Analysis Evaluation (2013)
Kumar, A., Sebastian, T.M.: Sentiment Analysis on Twitter. International Journal of Computer Science Issues (IJCSI)Â 9(4) (2012)
Lima, A., De Castro, L.: Automatic sentiment analysis of Twitter messages. In: Proceeding of the 2012 4th International Conference on Computational Aspects of Social Networks, pp. 52–57 (2012)
Logunov, A., Panchenko, V.: Characteristics and predictability of Twitter sentiment series. In: 19th International COngress on Modelling and Simulation, pp. 1617–1623 (2011)
Wang, X., Wei, F., Liu, X., et al.: Topic sentiment analysis in twitter: a graph-based hashtag sentiment classification approach. In: Proceedings of the 20th ACM International Conference on Information and Knowledge Management, pp. 1031–1040. ACM (2011)
Lin, J., Yang, A., Zhou, Y., et al.: Classification of Microblog Sentiment Based on Naive Bayesian. Computer Engineering & Science 9, 035 (2012)
Dong, Z., Dong, Q.: HowNet (2013), http://www.keenage.com [EB/OL]
National Taiwan University Sentiment Dictionary (NTUSD), Simplified Chinese sentiment polarity dictionary (2013), http://www.datatang.com/datares/go.aspx?dataid=601972
Zhang, W., Liu, J., Guo, X.: The Students Appraise Dictionary. Encyclopedia of China Publishing House (2004)
Li, J.: Experimental Study on Sentiment Classification of Chinese Reviews. Tsinghua University, Beijing (2008)
Liu, Q., Li, S.: Word similarity computing based on How-net. Computational Linguistics and Chinese Language Processing 7(2), 59–76 (2002)
Zhu, D.: Syntax Handouts. The Commercial Press (1982)
Liu, M.: Study of Chinese Negative Adverbs Origin and Diachronic Evolution, Hunan Normal University (2010)
Lin, H., Guo, S.: On the Characteristic, Range and Classification of Adverbs of Degree. Journal of Shanxi University 26(2), 71–74 (2003)
Shang, Y.: A semantic and pragmatic analysis of concurrence of degree adv. and bu. Journal of Zhoukou Normal University 23(3), 101–104 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Su, Z., Zhou, B., Li, A., Han, Y. (2014). Analysis on Chinese Microblog Sentiment Based on Syntax Parsing and Support Vector Machine. In: Han, W., Huang, Z., Hu, C., Zhang, H., Guo, L. (eds) Web Technologies and Applications. APWeb 2014. Lecture Notes in Computer Science, vol 8710. Springer, Cham. https://doi.org/10.1007/978-3-319-11119-3_10
Download citation
DOI: https://doi.org/10.1007/978-3-319-11119-3_10
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-11118-6
Online ISBN: 978-3-319-11119-3
eBook Packages: Computer ScienceComputer Science (R0)