Comparison of approaches to prioritized test generation for combinatorial interaction testing

Due to limited test resources, it is often necessary to prioritize and select test cases for a given system under test. Although test case prioritization is well studied and understood, its combination with test data generation is difficult and not completely solved yet. For example, the Classification Tree Method is a well established method for test data generation, however the application of prioritization techniques to it is a current research topic. We present an extension of the classification tree method that allows the generation of optimized test suites, containing test cases ordered according to their importance with respect to test goals. The presented algorithms are incorporated into the Classification Tree Editor and empirically evaluated on a set of benchmarks.