مدلسازی حالات و احساسات عمومی
Abstract
1. Introduction
2. Literature review
3. Research Methodology
4. Experimental results
5. Conclusion
Acknowledgment
References
Abstract
The development of online virtual communities has raised the importance in analyzing massive volume of text from websites and social networks. This research analyzed financial blogs and online news articles to develop a public mood dynamic prediction model for stock markets, referencing the perspectives of behavioral finance and the characteristics of online financial communities. This research applies big data and opinion mining approaches to the investors’ sentiment analysis in Taiwan. The proposed model was verified using experimental datasets from ChinaTimes.com, cnYES.com, Yahoo stock market news, and Google stock market news over an 18 month period. Empirical results indicate the big data analysis techniques to assess emotional content of commentary on current stock or financial issues can effectively forecast stock price movement.
Introduction
The rapid development of online communities and the mobile Internet have driven a rapid expansion in online news forums and discussions which potentially include data useful for investment decision making. Various approaches have been developed for analyzing “Big Data”, referred to as text, web and sentiment mining. Sentiment Mining is often referred to as opinion mining, sentiment analysis or subjectivity analysis. It is a form of textual analysis which automatically extracts words and sentences which appear with higher frequency and are potentially meaningful. “Sentiment” refers to contextualized attitudes, comments, and feelings, thus sentiment mining is designed to detect, extract, and analyze hidden sentiment or semantic orientation. Sentiment mining has been mostly applied to textual analysis social network content. In 2015, eMarketer found that 89% of US companies use social media as a marketing tool [1]. In 2013, RBC Capital found that the return on investment for advertising on Facebook is only slightly lower than that for top-ranked Google, and is ahead of Twitter, LinkedIn, Yahoo, AOL and other platforms [2], reflecting the significant marketing impact of social media. In addition, after U.S. Securities and Exchange Commission (SEC) officially allowed listed companies to disclose their earnings on social networks in 2013, the world’s leading data providers including Townsend Reuters and Bloomberg Data began to provide data analysis services for social network services. In 2014, the worlds’ largest social data provider GNIP noted that sentiment analysis social networks first began in 2010 [3]. The initial purpose of such activities was to allow companies to assess customer reaction to and satisfaction with their products and services. However, sentiment analysis of social networks has significant potential in other domains, such as the prediction of stock price movements.