• ISSN: 2301-3567 (Print), 2972-3981 (Online)
    • Abbreviated Title: J. Econ. Bus. Manag.
    • Frequency: Quarterly
    • DOI: 10.18178/JOEBM
    • Editor-in-Chief: Prof. Eunjin Hwang
    • Executive Editor: Ms. Fiona Chu
    • Abstracting/ Indexing:  CNKI, Google Scholar, Crossref
    • Article Processing Charge (APC): 500 USD
    • E-mail: joebm.editor@gmail.com
HOME > Archive > 2026 > Volume 14 Number 3 (2026) >
JOEBM 2026 Vol.14(3): 236-239
DOI: 10.18178/joebm.2026.14.3.947

Applications and Limitations of NLP-Based Financial News Sentiment Analysis for Short-Term Stock Market Prediction

Yifei Ren
Yifei Ren
Production arts for screen, University of arts London, London, UK
Email: y.ren0420241@arts.ac.uk

Manuscript received June 3, 2026; accepted August 12, 2026; published September 9, 2026.

Abstract—Financial news has become an important source of unstructured data for stock market analysis. Unlike structured indicators such as prices, trading volume and accounting ratios, news texts contain timely information about corporate events, macroeconomic conditions and investor expectations. With the development of Natural Language Processing (NLP), financial news sentiment can be extracted and converted into quantitative indicators for market analysis. This paper examines the applications and limitations of NLP-based financial news sentiment analysis in short-term stock market prediction. Based on existing literature, it reviews lexicon-based methods, traditional machine learning approaches, FinBERT, BERT-based models, and recent approaches based on Large Language Models (LLMs). The analysis shows that financial news sentiment can help capture investor expectations, explain short-term market reactions, and support predictions of returns, index movements, and trends. However, its value is limited by textual noise, information lag, semantic ambiguity, model misclassification, overfitting and changing market conditions. The paper argues that financial news sentiment should not be used as a standalone predictor. Instead, it is more appropriate as a supplementary signal to be combined with price, volume, financial indicators, model interpretability, and human verification.
 
Keywords—financial news, sentiment analysis, natural language processing, short-term stock market prediction, financial technology

Cite: Yifei Ren, "Applications and Limitations of NLP-Based Financial News Sentiment Analysis for Short-Term Stock Market Prediction," Journal of Economics, Business and Management, vol. 14, no. 3, pp. 236-239, 2026.

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).


PREVIOUS PAPER
The Value Game in a Polarizing Market: A Study on Pricing Strategies and Future Trends of Smart Refrigerators
NEXT PAPER
Research on Emergent Coordination and the Market Fragility in AI Trading Systems
Copyright © 2013-2026. Journal of Economics, Business and Management. Unless otherwise stated.
E-mail: joebm.editor@gmail.com