.png){kind=icon}
.png){kind=icon}
Return
A structural coupling network-based potential software risk identification method
.png){kind=icon}
PRE
AI
DOI：10.1016/j.jnca.2026.104487.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposes PSRI, a novel Structural Coupling Network-based method for identifying potential risk nodes in software systems. • Constructs a fine-grained, weighted directed SCN model that incorporates both function and variable entities and quantifies five key coupling relationships (Call, Data, Control, Return Value, and Variable Dependency) with precise strengths. • Defines two new risk metrics, VRI and DRI, based on a gravity model to assess a node’s Vulnerability Risk (influenced by others) and Diffusion Risk (influencing others), capturing asymmetric failure propagation. • Empirically validates the method on eight large-scale C/C++ open-source systems, demonstrating that PSRI significantly outperforms traditional complexity metrics and state-of-the-art key node identification algorithms in accurately identifying high-risk components.
Keywords:
PSRI
Structural Coupling Network
Risk Identification
Software Risk
Vulnerability Risk
Journal
IF:
8
Papers:
3.6K
Citations:
1.1W
Organization
No organization information available