Fuzzy Growing Map for Analyzing Cryptocurrency Market Trends

Articles in Press

Document Type : Research Paper

Authors

1 Department of Mangement and Economics, Islamic Azad University, Science and Research Branch, Tehran, Iran.

2 Department of Doctorate of Business Administration, Santamonica Training Academy, Hermes elm Gostar Institute, Isfahan, Iran.

3 Department of Business Management, Faculty of Management, Islamic Azad University, Firoozkooh, Iran.

4 Department of Management, UR. C., Islamic Azad University, Urmia, Iran.

10.22034/cmde.2025.69796.3445

Abstract

The cryptocurrency market is renowned for its rapid fluctuations and high volatility, making trend prediction a significant challenge due to the complexity and noise in the data. Traditional predictive models often struggle with such unstructured and high-dimensional datasets. This study introduces the Fuzzy Growing Map (FGM) as a cutting-edge approach to analyze cryptocurrency market trends. FGM combines fuzzy logic with dynamic growing maps to uncover hidden patterns and correlations in the data, making it especially suitable for the unpredictable and fast-paced nature of the cryptocurrency domain.
What sets FGM apart from conventional methods like neural networks, support vector machines, and regression models is its ability to dynamically adjust to new market changes in real-time}, without the need for time-consuming retraining. Unlike static models, FGM autonomously identifies key features from the input data and adapts continuously to emerging market conditions. When tested on historical cryptocurrency data, FGM demonstrated an impressive 92.3\% predictive accuracy, surpassing traditional models by 8-12\%.
The results reveal that FGM is not only more adaptive and responsive to market shifts but also more efficient in predicting trends, offering a significant improvement over existing methods. This makes FGM a groundbreaking tool for the future of cryptocurrency market forecasting, capable of tackling the challenges posed by the volatility and noise inherent in these markets.

Keywords

Main Subjects

Computational Methods for Differential Equations

Articles in Press, Accepted Manuscript
Available Online from 19 February 2026

Files

History

  • Receive Date: 21 October 2025
  • Revise Date: 11 November 2025
  • Accept Date: 11 November 2025

Share

How to cite

Statistics