A Comparative Study of Clustering Deviations for Black-Box Regression Test-Selection Techniques

Abstract

Regression test-selection techniques decrease the cost of regression testing by choosing a separation of an existing test suite to use in
retesting a customized program. Over the history, similarity based regression test-selection techniques have been described in the literature. This
paper aims to present a comparative study of present techniques of clustering deviations in black-box regression testing under the data mining
clustering and classification techniques that are in use in today's software engineering of verification and validation tasks. Number of
comparative study has been performed to evaluate the performance of predictive accuracy on the test cases and the outcome discloses that
Hierarchical Clustering (HC) and decision Tree outperforms having better performance other predictive methods like Simple K-means,
Randomized algorithms, are not performing well.

Keywords: Clustering, Regression testing, Black-box, Genetic algorithm.