Information propagation through a dependency network model

Author(s):  
Brian Drabble
2019 ◽  
Vol 30 (12) ◽  
pp. 2050005 ◽  
Author(s):  
Fuzhong Nian ◽  
Anhui Cong ◽  
Rendong Liu

This paper aims at the phenomenon of information selective propagation based on historical memory. A network model with memory strength and edge strength is established. The information propagation model with memory-clustering ability is designed with SIR model. And unsupervised learning is introduced to modify the performance. Based on the new network model, the core network and critical path that play a key role in the information propagation are found through the K-shell decomposition method. The research shows that the memory network contains an inertial channel for information propagation, it makes information propagation smooth. And information is selectively propagated in the new network, information is more inclined to propagate between nodes with powerful memory strength and close connections, in other words, people are more willing to propagate information to old friends who have been in contact for a long time instead of new friends.


2020 ◽  
Vol 2020 (1) ◽  
pp. 20005
Author(s):  
Heechun Kim ◽  
Jie Wu ◽  
Douglas A Schuler ◽  
Robert E. Hoskisson ◽  
Sung Hun Chung

2014 ◽  
Vol 602-605 ◽  
pp. 3355-3358
Author(s):  
Wang Xun Zhang ◽  
Yue Wang ◽  
Qun Li

There are often complex dependency relationships between systems in a system of systems (SoS), which present great challenges for SoS operability analysis. Functional Dependency Network Analysis offers the capability to evaluate the effect of both topology and of the possible degraded functioning of one or more systems on the operability of each node in the SoS. But there are some insufficiencies in the method, so here we give exact definition to self-effectiveness, which includes absolute self-effectiveness and relative self-effectiveness. Then according to the new definitions we show the detail calculation equations. Finally we we take the application of Guariniello and DeLaurentis’s five-node aerospace network for example, and compare the results with theirs. Comparative results show that their results are larger than the exact outcome, and our version give the authentic outcomes.


1991 ◽  
Vol 8 (1) ◽  
pp. 77-90
Author(s):  
W. Steven Demmy ◽  
Lawrence Briskin
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document