Software metrics enhance test data generation and productivity measurement

The adequate approach immensely desirable in test data generation as it is a difficult and costly task in system development process. Software metrics is successfully implemented to analyze the source code and to measure the performance of test data generation. In this work estimation of distribution algorithms employed the test data generation, which has been previously applied to the test data generation and using this data to compute the effort and productivity. The predictive capability of metrics is explored with respective outcome. It shows that the test data generation and productivity estimation based product metrics is also implemented effectively.