TY - GEN
T1 - A Novel Method for Discrete Evidence Fusion Based on Dijkstra Shortest Path Algorithm
AU - Liu, Bo
AU - Yang, Yang
AU - Deng, Xinyang
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/8
Y1 - 2020/8
N2 - Discrete evidence generally exists in the fields of expert systems, pattern recognition, and decision processing. How to solve the fusion of discrete evidence is an open question. In order to solve the problem that the number of calculations increases exponentially with the increase of the number of discrete evidences, a method for calculating the shortest path based on Dijkstra is proposed. First, the normalized discrete evidence is decomposed into the exact value of the evidence as far as possible, and then the Jousselme distance between the two pairs of evidence between the evidence groups is calculated. The Jousselme distance is designed to measure the conflict between the evidence. Then the Dijkstra shortest path algorithm is used to find the path with the largest and smallest Jousselme distance. The path here indicates the discrete evidence group with the largest fusion distance and smallest fusion distance. Then use Dempster combination to fuse the evidence after weighted average and give the interval value of BPA. Finally, two groups discrete evidence are given for verification to illustrate the rationality and effectiveness of the method.
AB - Discrete evidence generally exists in the fields of expert systems, pattern recognition, and decision processing. How to solve the fusion of discrete evidence is an open question. In order to solve the problem that the number of calculations increases exponentially with the increase of the number of discrete evidences, a method for calculating the shortest path based on Dijkstra is proposed. First, the normalized discrete evidence is decomposed into the exact value of the evidence as far as possible, and then the Jousselme distance between the two pairs of evidence between the evidence groups is calculated. The Jousselme distance is designed to measure the conflict between the evidence. Then the Dijkstra shortest path algorithm is used to find the path with the largest and smallest Jousselme distance. The path here indicates the discrete evidence group with the largest fusion distance and smallest fusion distance. Then use Dempster combination to fuse the evidence after weighted average and give the interval value of BPA. Finally, two groups discrete evidence are given for verification to illustrate the rationality and effectiveness of the method.
KW - Dempster combination
KW - Dijkstra algorithm
KW - Discrete evidence
KW - Jousselme distance
UR - http://www.scopus.com/inward/record.url?scp=85091589915&partnerID=8YFLogxK
U2 - 10.1109/CCDC49329.2020.9164188
DO - 10.1109/CCDC49329.2020.9164188
M3 - 会议稿件
AN - SCOPUS:85091589915
T3 - Proceedings of the 32nd Chinese Control and Decision Conference, CCDC 2020
SP - 3574
EP - 3579
BT - Proceedings of the 32nd Chinese Control and Decision Conference, CCDC 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 32nd Chinese Control and Decision Conference, CCDC 2020
Y2 - 22 August 2020 through 24 August 2020
ER -