Identifying important nodes affecting network security in complex networks

Citation: Liu, Y., Wang, J., He, H., Huang, G., & Shi, W. (2021). Identifying important nodes affecting network security in complex networks. International Journal of Distributed Sensor Networks, 17(2), 155014772199928.

Abstract: An important node identification algorithm based on an improved structural hole and K-shell decomposition algorithm
is proposed to identify important nodes that affect security in complex networks. We consider the global structure of a
network and propose a network security evaluation index of important nodes that is free of prior knowledge of network organization based on the degree of nodes and nearest neighborhood information. A node information control
ability index is proposed according to the structural hole characteristics of nodes. An algorithm ranks the importance of
nodes based on the above two indices and the nodes’ local propagation ability. The influence of nodes on network security and their own propagation ability are analyzed by experiments through the evaluation indices of network efficiency,
network maximum connectivity coefficient, and Kendall coefficient. Experimental results show that the proposed algorithm can improve the accuracy of important node identification; this analysis has applications in monitoring network
security.

Was the study experimental or non-experimental? Explain: The study was experimental. The abstract discusses experimental results that were found based on an algorithm that was tested as a solution to the problem.

Was the research qualitative or quantitative? Explain: I believe this study to be quantitative because in the analysis the authors use a table describing features measured by numbers. Equations were used in an effort to determine accuracy of the algorithm.

What was the population studied? Why do you say that? Since networks deal with machines, the population used in this study were three different domains for a simulation experiment. So the population here would be a network and the hosts (clients on a network) would be who is effected by the security vulnerability. The people utilizing the hosts would essentially be the “population” effected, as they would experience down time due to a security vulnerability taking down their network.

What sample was used for this study? Explain: Three datasets were used of various sizes to implement the “complex network” that was being looked for in the study. In these samples, nodes are linked to others and act as the center of information propagation and edges are the hosts that can’t connect to more than the center node. The first consisted of 34 nodes and 78 edges, formed by employees and members of a karate club. The second was a network of 62 dolphins containing 62 nodes and 159 edges. And the third was a network formed via political books sold during the presidential election in 2004 on Amazon.com, consisting of 105 nodes and 441 edges.

What was the method of measurement? With this being a quantitative study, I believe they used Ordinal measuring as the method. In this study, nodes and edges were deleted to see which were successfully ranked as most important to be identified as a security vulnerability. The algorithm would rank the nodes and the results would verify if it was correct or incorrect.

What was the method of analysis? In this study, the network efficiency, maximum connectivity coefficient, the Kendal coefficient. Theses all have equations connected to them to establish the analysis of the algorithm within each network in the sample.

What was the conclusion of the study? The ISHKS algorithm was more accurate and effective than two other popular algorithms that were tested and could be placed in a complex network environment.

Why is this study useful to you? Explain: This study is useful to me because I work in computer networks. I am pursuing the career of a Network Engineer and part of that is determining the best ways to keep my network secure from vulnerabilities. I currently work with a hospital that has many small networks attached so it is considered a complex environment. This study helps me understand how to determine issues of vulnerability within a complex network and is something I foresee me having to deal with in my near future.

What would be the next logical step in extending this study? I think the next logical step would be taking the better algorithm and putting it to the test against others in another complex environment. If you want to get away from testing algorithms that determine important nodes affecting security, then you could move on to software that best fights the vulnerabilities likely to breach your network security.