A Deng Entropy-Based Heuristic Method to Determine Discounting Coefficient in Dempster-Shafer Evidence Fusion

Area of Science:

• Information Fusion
• Data Science
• Artificial Intelligence

Background:

• Conflict management is essential in information fusion processes.
• Discounting methods are effective for handling conflicting data.
• Determining the appropriate discounting coefficient remains a challenge.

Purpose of the Study:

• To propose a novel heuristic method for determining the discounting coefficient in conflict management.
• To leverage Deng entropy and sigmoid function for coefficient calculation.
• To demonstrate the method's effectiveness and simplicity.

Main Methods:

• Utilizing Deng entropy to quantify evidence uncertainty.
• Employing a sigmoid function to map entropy to a discounting coefficient.
• Validating the method through numerical examples and a real-world application.

Main Results:

• The proposed heuristic method effectively determines discounting coefficients.
• The method shows promising performance compared to existing approaches.
• The approach is simple and easy to implement.

Conclusions:

• The Deng entropy and sigmoid function-based method offers an effective solution for conflict management in information fusion.
• This approach addresses the open issue of discounting coefficient determination.
• The method's practical applicability is confirmed through validation.