Abstract | ||
---|---|---|
Software metrics are collected in software development process and can be utilized to quantify software products, especially to predict software quality in the early stage of software life cycle. Data mining techniques have been applied to study software quality by analyzing software metrics. And clustering analysis, one of data mining techniques, has also been adopted to build software quality prediction models in the early period of software life cycle. However, not all kinds of software metrics are proper to be engaged in clustering analysis, and it is quite difficult to manually select them appropriately. Therefore, in this paper, based on the Genetic Algorithm (GA) and a new clustering method called Affinity Propagation (AP), we propose a novel strategy (GA-AP) to analyze software metrics for predicting software quality. Furthermore, we validate our new approach with two real-world software metrics datasets, and the experimental results show that GA-AP performs well in software metrics selection for clustering analysis. |
Year | Venue | Keywords |
---|---|---|
2008 | DMIN | affinity propagation,software metric,genetic algorithm |
Field | DocType | Citations |
Affinity propagation,Computer science,Software,Study software,Software development process,Artificial intelligence,Software metric,Cluster analysis,Software quality,Machine learning,Genetic algorithm | Conference | 3 |
PageRank | References | Authors |
0.43 | 17 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Bingbing Yang | 1 | 3 | 0.43 |
Xinyu Chen | 2 | 29 | 7.43 |
S. Xu | 3 | 14 | 3.64 |
Ping Guo | 4 | 601 | 85.05 |