TY - GEN
T1 - Research on Software Network Key Nodes Mining Methods Based on Complex Network
AU - Shan, Chun
AU - Wang, Peng
AU - Hu, Changzhen
AU - Gao, Xianwei
AU - Mei, Shanshan
N1 - Publisher Copyright:
© Springer Nature Singapore Pte Ltd 2020.
PY - 2020
Y1 - 2020
N2 - Complex software system will bring a lot of software security problems, it is very meaningful to know how to more accurately abstract the software network model from the software system and efficiently find the key nodes in the software network. This research takes open source software as the research object, constructs a directed network model of software system, proposes a new weight calculation method, adds weights to the model to form a directed weighted network, and then regards the software network as a complex network. The defect mining and defect propagation cost are two node mining methods related to the weight and degree of the node. At the same time, the PageRank algorithm is improved to mine the key nodes. Finally, the robustness of the software system execution network model is carried out by different attack methods. The evaluation, through experimental verification and comparison, shows that the mining method proposed in this study can more accurately and efficiently mine key nodes in the software system.
AB - Complex software system will bring a lot of software security problems, it is very meaningful to know how to more accurately abstract the software network model from the software system and efficiently find the key nodes in the software network. This research takes open source software as the research object, constructs a directed network model of software system, proposes a new weight calculation method, adds weights to the model to form a directed weighted network, and then regards the software network as a complex network. The defect mining and defect propagation cost are two node mining methods related to the weight and degree of the node. At the same time, the PageRank algorithm is improved to mine the key nodes. Finally, the robustness of the software system execution network model is carried out by different attack methods. The evaluation, through experimental verification and comparison, shows that the mining method proposed in this study can more accurately and efficiently mine key nodes in the software system.
KW - Defect probability
KW - Defect propagation cost
KW - Improved PageRank algorithm
KW - Key nodes
KW - Software security
UR - http://www.scopus.com/inward/record.url?scp=85081627597&partnerID=8YFLogxK
U2 - 10.1007/978-981-15-3418-8_14
DO - 10.1007/978-981-15-3418-8_14
M3 - Conference contribution
AN - SCOPUS:85081627597
SN - 9789811534171
T3 - Communications in Computer and Information Science
SP - 193
EP - 205
BT - Trusted Computing and Information Security - 13th Chinese Conference, CTCIS 2019, Revised Selected Papers
A2 - Han, Weili
A2 - Zhu, Liehuang
A2 - Yan, Fei
PB - Springer
T2 - 13th Chinese Conference on Trusted Computing and Information Security, CTCIS 2019
Y2 - 24 October 2019 through 27 October 2019
ER -