Sentiment Classification from Social Media for Stock Prediction with Data Mining
Natassya Afdalena Triany1, Sani M. Isa2
1Natassya Afdalena Triany*, Department of information technology from BINUS University Indonesia.
2Sani Muhamad Isa, Lecturer and Researcher in the Department of Computer Science, BINUS University Indonesia.
Manuscript received on February 10, 2020. | Revised Manuscript received on February 21, 2020. | Manuscript published on March 10, 2020. | PP: 155-161 | Volume-9 Issue-5, March 2020. | Retrieval Number: E2226039520/2020©BEIESP | DOI: 10.35940/ijitee.E2226.039520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Social media currently plays an important role as a means of exchanging information. Through social media, information is obtained that can be used to see people’s sentiments about a product or an event. Social media is a viable option to attract public sentiment through a method called sentiment analysis. The thing done is attracting sentiment from internet users through the posts made. In this way, sentiment data can be collected quickly and easily. Current economic behavior has proven that financial decisions are driven significantly by sentiment. The level of collective optimism or pessimism in society can influence investor decisions. Sentiment can also be interpreted as something that is felt by someone, both positive and negative. Sentiments and perceptions are psychological constructs and therefore difficult to measure in the analysis. This study focuses on sentiment analysis of information obtained from Twitter about stocks. For sentiment classification process ensemble methods of Naïve Bayes and SVM is used. Sentiment results are classified as positive or negative. We are expecting to see if there is connection between sentiment analysis from social media in predicting movement of IHSG stock price. As a result, we obtained strong correlation with coefficient of correlation r= 0.56609.
Keywords: Sentiment Analysis, Twitter, Stock Index, Classification, Data Mining.
Scope of the Article: Data Mining and Warehousing