TY - JOUR
T1 - A Novel Algorithm for Identifying Key Function Nodes in Software Network Based on Evidence Theory
AU - Wang, Qian
AU - Shan, Chun
AU - Zhao, Xiaolin
AU - Dong, Jun
AU - Ren, Jiadong
AU - Liu, Jiaxin
N1 - Publisher Copyright:
© 2019 World Scientific Publishing Company.
PY - 2019/3/1
Y1 - 2019/3/1
N2 - In a software network system, it is of great significance to identify key functions for software fault detection and maintenance. In order to better understand the characteristics and internal structure of software, a key Node Discovery algorithm based on Evidence Theory called NDET is proposed in this paper. First, the software complex network model is constructed according to the execution process of the software. Based on the Dempster-Shafer evidence theory (D-S evidence theory), the discernment frame is formed, the maximum and minimum values of the network degree and strength are determined. Second, the Basic Probability Assignment (BPA) of each node degree is calculated by considering the node degree distribution ratio value. Third, based on Dempster's rule of combination, the evidential centrality of the node itself and the fluctuation value of the node influenced by neighbor nodes are considered for the key measurement. Finally, by using the Susceptible-Infected-Recovered (SIR) model to simulate the spreading process on real software networks, the performance of NDET is evaluated. Experiment results verify the validity and accuracy of NDET for identifying key function nodes in software.
AB - In a software network system, it is of great significance to identify key functions for software fault detection and maintenance. In order to better understand the characteristics and internal structure of software, a key Node Discovery algorithm based on Evidence Theory called NDET is proposed in this paper. First, the software complex network model is constructed according to the execution process of the software. Based on the Dempster-Shafer evidence theory (D-S evidence theory), the discernment frame is formed, the maximum and minimum values of the network degree and strength are determined. Second, the Basic Probability Assignment (BPA) of each node degree is calculated by considering the node degree distribution ratio value. Third, based on Dempster's rule of combination, the evidential centrality of the node itself and the fluctuation value of the node influenced by neighbor nodes are considered for the key measurement. Finally, by using the Susceptible-Infected-Recovered (SIR) model to simulate the spreading process on real software networks, the performance of NDET is evaluated. Experiment results verify the validity and accuracy of NDET for identifying key function nodes in software.
KW - Dempster-Shafer evidence theory
KW - Software structure
KW - complex network
KW - key function
UR - http://www.scopus.com/inward/record.url?scp=85063506374&partnerID=8YFLogxK
U2 - 10.1142/S0218194019500189
DO - 10.1142/S0218194019500189
M3 - Article
AN - SCOPUS:85063506374
SN - 0218-1940
VL - 29
JO - International Journal of Software Engineering and Knowledge Engineering
JF - International Journal of Software Engineering and Knowledge Engineering
IS - 3
M1 - 00189
ER -