The purpose of this project is to improve the automation of testing activities related to software systems and as a consequence the results of the test activities.
Examples of improvements are: reduced time to prepare and run tests, increase in the discovery of undetected faults and reduction in the amount of executed, unnecessary tests. The approach in this project is to apply semantic technologies and ontology engineering. The main purpose is to create a method for deriving test case data from an ontology representing the specification and domain for a software system. The ontology will be utilised to generate test data through inference rules, formalising the knowledge of experienced testers in the domain. Furthermore, this approach to ontology-based testing will be combined with evolutionary testing, e.g. genetic algorithms. Specifically, the fitness function will be defined through the domain relevance of generated test data as well as mutations which will partly be performed with the help of the domain ontology.
The most important industrial needs addressed in this project are to prepare test case data without actual test execution, reduce complexity of searching for test case data, ensure diversity of generated test cases and to reduce the amount of resources required for test case preparation.
The project work includes four major activities:
Vladimir Tarasov, Associate Professor
Content updated 2017-03-08