Scholarlink

<p>Soft computing techniques in fuzzy database mining have paved a way for solving complex data analysis problems related to financial risk forecasting. In this paper, an ensemble approach for financial risk forecasting is proposed for financial risk forecasting. The proposed approach utilizes collection of soft computing techniques such as fuzzy logic (FL) to handle uncertain financial data, convolutional neural network (CNN) to tune the obtained fuzzy parameters followed by optimization of neural network model using genetic algorithm (GA). The proposed approach can model intricate financial scenarios where classical statistical approaches face difficulties. GA optimizes Fuzzy Rules and Membership functions while CNN help make predictions more accurate. The results reveal that the forecasting accuracy evidently improves as compared to those using traditional methods, illustrating promising applications of integrating soft computing within financial risk management.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A hybrid soft computing approach to analyze and forecast financial risks

  • 期刊论文

  • 发表日期：

    2025年6月2日

查看全文
  • M. B. Abhishek,
  • Rajesh Gadipuuri,
  • Surjeet,
  • Hiteshwari Sabrol,
  • Vipul Dalal,
  • Archna Ratnaparkhi,
  • Ramesh Krishnamaneni

摘要

Soft computing techniques in fuzzy database mining have paved a way for solving complex data analysis problems related to financial risk forecasting. In this paper, an ensemble approach for financial risk forecasting is proposed for financial risk forecasting. The proposed approach utilizes collection of soft computing techniques such as fuzzy logic (FL) to handle uncertain financial data, convolutional neural network (CNN) to tune the obtained fuzzy parameters followed by optimization of neural network model using genetic algorithm (GA). The proposed approach can model intricate financial scenarios where classical statistical approaches face difficulties. GA optimizes Fuzzy Rules and Membership functions while CNN help make predictions more accurate. The results reveal that the forecasting accuracy evidently improves as compared to those using traditional methods, illustrating promising applications of integrating soft computing within financial risk management.