Software Fault Feature Clustering Algorithm Based on Sequence Pattern

Authors:
Jiadong Ren;Changzhen Hu;Kunsheng Wang;Dongmei Zhang
Affiliations:
College of Information Science and Engineering, Yanshan University, Qinhuangdao, China 066004 and School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China 100081;School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China 100081;China Aerospace Engineering Consultation Center, Beijing, China 100037;College of Information Science and Engineering, Yanshan University, Qinhuangdao, China 066004
Venue:
WISM '09 Proceedings of the International Conference on Web Information Systems and Mining
Year:
2009

Citing 6
Cited 0

Knowledge for Software Security

IEEE Security and Privacy
SEQOPTICS: A Protein Sequence Clustering Method

IMSCCS '06 Proceedings of the First International Multi-Symposiums on Computer and Computational Sciences - Volume 1 (IMSCCS'06) - Volume 01
A Software Fault Tree Metric

ICSM '06 Proceedings of the 22nd IEEE International Conference on Software Maintenance
A Metrics Framework to Drive Application Security Improvement

IEEE Security and Privacy
Software fault tree and coloured Petri net based specification, design and implementation of agent-based intrusion detection systems

International Journal of Information and Computer Security
Software Security; A Vulnerability Activity Revisit

ARES '08 Proceedings of the 2008 Third International Conference on Availability, Reliability and Security

Quantified Score

Hi-index	0.00

Visualization

Abstract

Software fault feature analysis has been the important part of software security property analysis and modeling. In this paper, a software fault feature clustering algorithm based on sequence pattern (SFFCSP) is proposed. In SFFCSP, Fault feature matrix is defined to store the relation between the fault feature and the existing sequence pattern. The optimal number of clusters is determined through computing the improved silhouette of fault feature matrix row vector, which corresponds to the software fault feature. In the agglomerative hierarchical clustering phase, entropy is considered as the similarity metric. In order to improve the time complexity of the software fault feature analysis, the fault features of the software to be analyzed are matched to each centroid of clustering results. Experimental results show that SFFCSP has better clustering accuracy and lower time complexity compared with the SEQOPTICS.