Exploring Social Media's Role in Predicting Stock Market Trends
This study analyzes tweets from the official Twitter accounts of NHK News and Nikkei to incorporate sentiment data into a predictive model for the Nikkei Stock Average.Adding sentiment data improved the R score from 45.1% to a maximum of 70.5%, indicating the potential of SNS data in forecasting social indicators.However, no strong correlation between sentiment data and stock prices was observed.Challenges include the short data collection period and the difficulty of sentiment analysis in Japanese.Future work should focus on employing more effective methods for extracting sentiment.
Authors
- Tsutomu Ito (ORCID: https://orcid.org/0000-0001-6033-2793)
- Amane Takei (ORCID: https://orcid.org/0000-0001-9014-8727)
- Makoto Sakamoto (ORCID: https://orcid.org/0000-0003-0383-1304)
- Satoshi Ikeda (ORCID: https://orcid.org/0000-0002-6946-1601)
- Minoru Ide
- Takao Ito
- Masatoshi Beppu
- Seita Nagashima
Institutions
- Hiroshima University (JP)
- University of Miyazaki (JP)
- National Institute of Technology, Ube College (JP)
- JTEC Corporation (Japan)
Publication Details
- Journal
- Proceedings of International Conference on Artificial Life and Robotics
- Published
- 2025-02-13
- DOI
- https://doi.org/10.5954/icarob.2025.os23-4
- Primary Topic
- Stock Market Forecasting Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Japan Society for the Promotion of Science