Volume 37, Number 5, October 2019
|Page(s)||897 - 902|
|Published online||14 January 2020|
K-Terminal Network Permutation Importance Measure Based on Mixture C-Spectrum
School of Science, Lanzhou University of Technology, Lanzhou 730050, China
2 School of Mechatronics, Northwestern Polytechnical University, Xi'an 710072, China
The construction spectrum(C-spectrum) is often used to exploit the network reliability and importance measure. It depends only on the network structure and hence called structure invariant. Importance measure can be used to quantify the criticality of edge within a network. This paper aim at generalizing the traditional permutation importance measure to accommodate the case of K-terminal network in which all the edges fail with independent and equal probability. A concept for mixture C-spectrum is introduced to evaluate the permutation importance measure of edges. It is proved that the rankings according to the permutation importance measure depend only on the network structure through the mixture C-spectrum when the network has special structure or the reliability of edge is sufficient large. Finally, numerical experiment show that the Monte Carlo algorithm based on the mixture C-spectrum can be efficiently used to evaluate the permutation importance measure.
Key words: K-terminal network / permutation importance measure / mixture C-spectrum / Monte Carlo
关键字 : K-终端网络 / 置换重要度 / 混合C-谱 / 蒙特卡罗
© 2019 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.