Multi-angle evaluations of test cases based on dynamic analysis

Abstract

This paper presents dynamic analysis of test cases. By software mining, we get dynamic call tree to reproduce the dynamic function calling relations of test cases and static call graph to describe the static calling relations. Based on graph analysis, we define some related testing models to evaluate the test cases with the execution of software. Compared with the models of evaluating test cases in static analysis, the models given in this paper can be used on large-scale software systems and the quantization can be completed automatically. Experiments prove that these models of dynamic analysis have an excellent performance in improving testing efficiency and also build a foundation of quantization for the management, selection, evaluation of capability to find software defects of test cases. Even more critical is that they can indicate the directions of improvement and management of the test for testers.