Innovative Sentiment Analysis and Prediction of Stock Price Using FinBERT, GPT-4 and Logistic Regression: A Data-Driven Approach

被引:1
|
作者
Shobayo, Olamilekan [1 ]
Adeyemi-Longe, Sidikat [1 ]
Popoola, Olusogo [1 ]
Ogunleye, Bayode [2 ]
机构
[1] Sheffield Hallam Univ, Sch Comp & Digital Technol, Sheffield S1 2NU, England
[2] Univ Brighton, Dept Comp & Math, Brighton BN2 4GJ, England
关键词
FinBERT model; logistic regression; FinBERT; Optuna; time series cross-validation;
D O I
10.3390/bdcc8110143
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This study explores the comparative performance of cutting-edge AI models, i.e., Finaance Bidirectional Encoder representations from Transsformers (FinBERT), Generatice Pre-trained Transformer GPT-4, and Logistic Regression, for sentiment analysis and stock index prediction using financial news and the NGX All-Share Index data label. By leveraging advanced natural language processing models like GPT-4 and FinBERT, alongside a traditional machine learning model, Logistic Regression, we aim to classify market sentiment, generate sentiment scores, and predict market price movements. This research highlights global AI advancements in stock markets, showcasing how state-of-the-art language models can contribute to understanding complex financial data. The models were assessed using metrics such as accuracy, precision, recall, F1 score, and ROC AUC. Results indicate that Logistic Regression outperformed the more computationally intensive FinBERT and predefined approach of versatile GPT-4, with an accuracy of 81.83% and a ROC AUC of 89.76%. The 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 resource-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 the practical applications of AI approaches in stock market prediction and presents Logistic Regression as the most efficient model for this task, with FinBERT and GPT-4 representing emerging tools with potential for future exploration and innovation in AI-driven financial analytics.
引用
收藏
页数:21
相关论文
共 50 条
  • [41] Data-driven approach to machine condition prognosis using least square regression tree
    Van Tung Tran
    Bo-Suk Yang
    Journal of Mechanical Science and Technology, 2009, 23 : 1468 - 1475
  • [42] A Data-Driven Approach to Spacecraft Attitude Control Using Support Vector Regression (SVR)
    Mahayana, Dimitri
    IEEE ACCESS, 2024, 12 : 177896 - 177910
  • [43] Prediction of Railway Track Condition for Preventive Maintenance by Using a Data-Driven Approach
    Vale, Cecilia
    Simoes, Maria Lurdes
    INFRASTRUCTURES, 2022, 7 (03)
  • [44] Data-driven approach to machine condition prognosis using least square regression tree
    Tran, Van Tung
    Yang, Bo-Suk
    JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY, 2009, 23 (05) : 1468 - 1475
  • [45] Urban building energy performance prediction and retrofit analysis using data-driven machine learning approach
    Ali, Usman
    Bano, Sobia
    Shamsi, Mohammad Haris
    Sood, Divyanshu
    Hoare, Cathal
    Zuo, Wangda
    Hewitt, Neil
    O'Donnell, James
    ENERGY AND BUILDINGS, 2024, 303
  • [46] The future capacity prediction using a hybrid data-driven approach and aging analysis of liquid metal batteries
    Shi, Qionglin
    Zhao, Lin
    Zhang, E.
    Xia, Junyi
    Li, Haomiao
    Wang, Kangli
    Jiang, Kai
    JOURNAL OF ENERGY STORAGE, 2023, 67
  • [47] Urban building energy performance prediction and retrofit analysis using data-driven machine learning approach
    Ali, Usman
    Bano, Sobia
    Shamsi, Mohammad Haris
    Sood, Divyanshu
    Hoare, Cathal
    Zuo, Wangda
    Hewitt, Neil
    O'Donnell, James
    Energy and Buildings, 2024, 303
  • [48] Institutional investors’ distraction and the quality of accounting information disclosure: A data-driven approach using multiple regression analysis
    Si Fang
    Chongyan Cao
    Journal of Data, Information and Management, 2025, 7 (1): : 21 - 36
  • [49] Event-driven sentiment analysis for stock prediction using constructed domain-specific Chinese financial sentiment lexicon Take the stock price of Haitian Flavouring & Food Company Ltd. as an example
    Zhu, Yanlin
    Zhang, Ming
    Chen, Jiazhen
    Fan, Longhao
    2023 THE 6TH INTERNATIONAL CONFERENCE ON ROBOT SYSTEMS AND APPLICATIONS, ICRSA 2023, 2023, : 149 - 156
  • [50] Distributions of fatigue damage from data-driven strain prediction using Gaussian process regression
    Gibson, Samuel J.
    Rogers, Timothy J.
    Cross, Elizabeth J.
    STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL, 2023, 22 (05): : 3065 - 3076