Applying Swarm Ensemble Clustering Technique for Fault Prediction Using Software Metrics
| Autor: | Fabricio dos R.N. Guimaraes, Rodrigo A. Coelho, Ahmed A. A. Esmin |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
Computer science
business.industry Correlation clustering Particle swarm optimization Swarm behaviour Pattern recognition computer.software_genre Biclustering ComputingMethodologies_PATTERNRECOGNITION CURE data clustering algorithm Robustness (computer science) Consensus clustering Canopy clustering algorithm Data mining Artificial intelligence Multi-swarm optimization Cluster analysis business computer |
| Zdroj: | ICMLA |
| DOI: | 10.1109/icmla.2014.63 |
| Popis: | Number of defects remaining in a system provides an insight into the quality of the system. Defect detection systems predict defects by using software metrics and data mining techniques. Clustering analysis is adopted to build the software defect prediction models. Cluster ensembles have emerged as a prominent method for improving robustness, stability and accuracy of clustering solutions. The clustering ensembles combine multiple partitions generated by different clustering algorithms into a single clustering solution. In this paper, the clustering ensemble using Particle Swarm Optimization algorithm (PSO) solution is proposed to improve the prediction quality. An empirical study shows that the PSO can be a good choice to build defect prediction software models. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|