RE-COST: REducing the Cost of Oracles for Software Testing

Lead Research Organisation: University of Sheffield
Department Name: Computer Science

Abstract

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
 
Description 1) Readability of automatically-generated test inputs is key when it comes to users of automatic test input generation tools in practice.

2) The readability of automatically-generated string test inputs can be improved, lowering the time needed for humans to check the corresponding outputs to those inputs, in particular through:

a) strings sourced from the Internet.

b) through combining meta-heuristic algorithms with statistical natural language models

3) That the costs of generating test data can be reduced by combining symbolic execution with search based test data generation

4) That the success rate of test data generation for Java programs may be improved when Memetic algorithms are used.
Exploitation Route The research can be used by companies that develop software testing tools. The test data generated by those tools may be made more realistic and human-readable, thus reducing costs of testing for the tool users.
Sectors Digital/Communication/Information Technologies (including Software)