AI-Powered Sentiment Analysis for Stock Market Prediction in Renewable Energy
DOI:
https://doi.org/10.26821/ijshre.14.09.2026.140904Keywords:
sentiment analysis, machine learning, stock prediction, renewable energy, NLP, trading strategyAbstract
This paper presents a joint effort between Natural Language Processing (NLP) and Market Technical Analysis (MTA), powered by Artificial Intelligence (AI) to predict stock price movements in renewable energy companies. The A.I. system leverages FinBERT (a domain-specific transformer model) to extract sentiment, ticker relevance, and other business event types from over 850 articles about five different renewable energy stocks (Tesla, Enphase Energy, SolarEdge, NextEra Energy, Brookfield Renewable). It aggregates sentiment signals over time, introduces decay weights, and merges them with other MTA features such as volatility and recent return to predict price movements of each of the five renewable energy stocks from 1 to 5 trading day(s) in the future. A rolling time-series validation process guarantees temporal integrity and helps eliminate look-ahead bias in the A.I. system results. The combined model achieves 62% accuracy for directional predictions of each stock price movement on the next trading day, far surpassing the accuracy of the two standalone models (48% accuracy for sentiment only and 52% accuracy for market only). The A.I. identified the most significant event types associated with each of the five stocks' predicted price movement: capacity announcements, subsidy changes, and supply chain disruptions. This demonstration illustrates the immense potential for applying multi-modal machine learning to help improve financial forecasting across other highly regulated sectors as well.
References
Araci, Dogu. "FinBERT: Financial Sentiment Analysis with Pre-Trained Language Models." arXiv Preprint arXiv:1908.10063, 2019, pp. 1-7.
Chen, Henry, et al. "The Role of Earnings Guidance in Firm Valuation: Evidence from the 2008–2009 Financial Crisis." Financial Management, vol. 47, no. 3, 2018, pp. 731-753.
Fama, Eugene F., and Kenneth R. French. "The Cross-Section of Expected Stock Returns." The Journal of Finance, vol. 47, no. 2, 1992, pp. 427-465.
Gentzkow, Matthew, and Jesse M. Shapiro. "What Drives Media Slant? Evidence from US Newspapers." Econometrica, vol. 78, no. 1, 2010, pp. 35-71.
Hastie, Trevor, et al. The Elements of Statistical Learning: Data Mining, Inference, and Prediction. 2nd ed., Springer-Verlag, 2009.
Hoseinzade, Ehsan, and Saman Haratizadeh. "CNNpred: CNN-based Stock Market Prediction Using a Diverse Set of Variables." Expert Systems with Applications, vol. 121, 2019, pp. 113-123.
Huang, Allen H., et al. "Artificial Intelligence Capabilities of Auditors: Perceptions from U.S. Audit Practitioners." Journal of Information Systems, vol. 34, no. 2, 2020, pp. 23-44.
Kaplan, Steven N., and Ilya Vaicunt. "The Emergence of Risk-Based Management at the Margin." Journal of Financial Economics, vol. 115, no. 2, 2015, pp. 265-280.
Krauss, Christopher, et al. "Deep Neural Networks, Gradient-Based Learning Methods, and Adversarial Examples." European Journal of Operational Research, vol. 259, no. 3, 2017, pp. 945-959.
Liu, Bing. Sentiment Analysis and Opinion Mining. Morgan & Claypool Publishers, 2020.
Mittal, Anshul, and Arpit Goel. "Stock Prediction Using Twitter Sentiment Analysis." Stanford CS229 Machine Learning Class Project, Stanford University, 2012.
Nassirtoussi, Arman K., et al. "Text Mining for Stock Market Prediction: A Systematic Review." Expert Systems with Applications, vol. 41, no. 15, 2014, pp. 7653-7670.
Poterba, James M., and Lawrence H. Summers. "Mean Reversion in Stock Prices: Evidence and Implications." Journal of Financial Economics, vol. 22, no. 1, 1988, pp. 27-59.
Tetlock, Paul C. "Giving Content to Investor Sentiment: The Role of Media in Finance." The Journal of Finance, vol. 62, no. 3, 2007, pp. 1139-1168.
Ye, Hao, et al. "Stock Price Movement Prediction with Convolutional Neural Network." Proceedings of the International Conference on Machine Learning and Cybernetics (ICMLC), IEEE, 2016, pp. 93-99.
