earticle

논문검색

An Efficient Combination Method of Conflict Evidences

초록

영어

Evidence theory is an effective method for uncertainty reasoning, which is widely used in areas like expert system, artificial intelligence, pattern recognition and system decision. But traditional DS combination rule will produce the result contrary to intuition on the condition of high confliction. To solve the problem, this paper proposes a modified method based on the option of distance function and correction of support degree. Firstly, it introduces the Minkowsky distance as distance function of evidences and finds the support degree of each evidence in system, then corrects the support degree on the basic of its distribution, finally, it gets the weighted average of evidence by the normalized evidence credibility, and uses the DS combination rule to synthesis evidences. The simulation results demonstrate the effectiveness and reliability of modified method.

목차

Abstract
 1. Introduction
 2. Review of Evidence Theory
 3. Problems and Classical Solutions
  3.1. Problems
  3.2. Classical Solutions
 4. Modified Combination Rule
 5. Simulation
  5.1. Simulation 1
  5.2. Simulation 2
 6. Conclusion
 References

저자정보

  • Yibing Li College of Information and Communication Engineering Harbin Engineering University Harbin 150001, China
  • Jie Chen College of Information and Communication Engineering Harbin Engineering University Harbin 150001, China
  • Yun Lin College of Information and Communication Engineering Harbin Engineering University Harbin 150001, China

참고문헌

자료제공 : 네이버학술정보

    함께 이용한 논문

      ※ 원문제공기관과의 협약기간이 종료되어 열람이 제한될 수 있습니다.

      0개의 논문이 장바구니에 담겼습니다.