Real-time fault detection approach of software under big data environment
- 10.2991/amcce-15.2015.215How to use a DOI?
- K-means clustering algorithm; feature extraction; real-time software fault detection
For large data environment, the traditional K-means clustering algorithm flaw in the software in real-time fault detection process, we propose a K-means clustering software intelligent real-time fault detection method is improved. The K-means clustering algorithm and particle swarm optimization combined during the iterative process, the combination of K-means to optimize the upcoming PSO algorithm offspring individuals use K-means clustering to obtain local optimal solution calculates and uses these individuals continue to participate in an iterative process, so that the algorithm can improve the convergence speed, avoid falling into local optimal solution, to obtain accurate software fault signal characteristics. Experimental results show that the use of K-means tilt feature extraction software, intelligent real-time fault detection algorithm can effectively improve the accuracy of fault detection, and achieved satisfactory results. On the condition of big data, traditional K-mean cluster algorithm showed some flaws of software fault detection in real-time way. In this research, an improved k-mean cluster method was proposed for the purpose of testing software fault in intellectual and real-time way. This improved method combined k-mean cluster algorithm with Particle cluster algorithm. K-mean cluster were used in the optimizing procedure among the iterative process. Individuals from Particle cluster algorithm were computed by using K-mean algorithm and obtained the locally optimal solution, and then these individuals continue to participate in the iterative processing. This improved converging rate of the algorithm, avoided falling into the locally optimal solution and finally got the accurate features of software fault signal. Results showed that this testing method with using the tilt feature of k-mean algorithm improved detection accuracy effectively.
- © 2015, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Xianrui Jian PY - 2015/04 DA - 2015/04 TI - Real-time fault detection approach of software under big data environment BT - Proceedings of the 2015 International Conference on Automation, Mechanical Control and Computational Engineering PB - Atlantis Press SN - 1951-6851 UR - https://doi.org/10.2991/amcce-15.2015.215 DO - 10.2991/amcce-15.2015.215 ID - Jian2015/04 ER -