Simulation experiment result forecast method based on support vector machine (SVM) is put forward and elaborated. Two-class and multiclass forecast experiments are successfully performed. And SVMs are proved to be powerful in solving the high time cost problem occurring in the exploratory simulation analysis.
Network-of-Networks (NoN) is the latest developments of Complex Network, which is suitable for researching problem of System-of-Systems(SoS). We put forward a conceptual model of military SoS based on NoN, the model can describe various dynamic and static factors of SoS effectively. The SoS is mapped into physical network, relationship network and… (More)
Content distribution network is an acceleration technology for Internet content. The content was pushed or pulled to the edge of the Internet. So the user can access the content closely. In this paper, the contents routing problem in content distribution network was researched. There are two kind content routing method, file-level routing strategies and… (More)
This paper applies dynamics model to understand the origin of dominant behaviors in social networks that generally lead to emergence of personage, i.e. few leaders with overwhelming popular support. We commentate the notion of behaviors selection networks and introduce an improved social behaviors selection model for evolutionary dynamics of behaviors in… (More)
A series of support vector machine (SVM) forecast experiments are carried out to reveal the relation between the SVM training sample size and SVM correct forecast ratio for simulation experiment results. Experiment results show that the SVM correct forecast ratio increases to some extent with the number of training samples becoming more and then keeps… (More)