Innovative Sentiment Analysis and Prediction of Stock Price Using FinBERT, GPT-4 and Logistic Regression: A Data-Driven Approach
Year: 2024
Type: article
Abstract: This study explores the comparative performance of FinBERT, GPT-4, and Logistic Regression for sentiment analysis and stock index prediction using the NGX All-Share Index dataset. By leveraging advanced language models like GPT-4 and FinBERT, alongside a traditional machine learning model, Logistic Regression, we aim to classify market sentiment, generate sentiment score, and predict market price movements. The models were assessed using metrics such as accuracy, precision, recall, F1 score, and ROC AUC. Results indicate that Logistic Regression outperformed both FinBERT and GPT-4, with an accuracy of 81.83% and a ROC AUC of 89.76%. GPT-4 predefined approach exhibited a lower accuracy of 54.19% but demonstrated strong potential in handling complex data. FinBERT, while offering more sophisticated analysis, was computationally demanding and yielded a moderate performance. Hyperparameter optimization using Optuna and cross-validation techniques ensured the robustness of the models. This study highlights the strengths and limitations of these approaches in stock market prediction and presents Logistic Regression as the most efficient model for this task, with FinBERT and GPT-4 showing promise for future exploration.
Source: Big Data and Cognitive Computing
DOI: 10.3390/bdcc8110143
Institutions: Sheffield Hallam University, University of Brighton
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $2,046
OpenAlex ID: W4403765241