AUTOMATIC TEST DATA GENERATION TOOL FOR LARGE-SCALE SOFTWARE SYSTEMS

We propose the development of an Automatic Testing Data Generation Tool for Large Scale Commercial Applications. The tool will use a novel hybrid technique of symbolic execution and dynamic flow analysis. The proposed approach will be able to generate different forms of input data for large commercial applications. The technique can be applied to generating input data for distributed and imbedded applications as well as traditional applications. The tool can be used in mutation testing for the automatic generation of input cases, and will significantly improve the applicability of this advanced method. During Phase I we propose building a working prototype of the tool, and applying the tool to a set of commercial applications.