Date of Award

3-2023

Document Type

Dissertation

Degree Name

Ph.D.

Organizational Unit

Daniel Felix Ritchie School of Engineering and Computer Science, Computer Science

First Advisor

Kerstin Haring

Second Advisor

Anneliese Andrews

Third Advisor

Chris GauthierDickey

Fourth Advisor

Chip Reichardt

Keywords

Mobile applications, Regression testing

Abstract

Mobile Applications have been widely used in recent years daily all over the world and are essential in our personal lives and at work. Because Mobile Applications update frequently, it is important that developers perform regression testing to ensure their quality. In addition, the Mobile Applications market has been growing rapidly, allowing anyone to write and publish an application without appropriate validation. A need for regression testing has arisen with the growth of different Mobile Apps and the added functionalities and complexities. In this dissertation, we adapted the FSMWeb [14] approach for selective regression testing to allow for selective regression testing of Mobile Apps. We applied rules to classify the original set of tests of the Mobile App into obsolete, retestable, and reusable tests based on the types of changes to the model of Mobile Apps. New tests are added to cover portions that have not been tested.

As regression test suites change, we want to ensure that required tests are included to satisfy testing criteria, but also that redundant tests are removed, so as not to bloat the regression tests suite. In the dissertation, we developed a test case minimization approach for FSMApp, based on concept analysis that removes redundant test cases.

Next, we proposed an approach to prioritize test cases for Mobile Apps. Naturally, it is desirable to select those test cases that are most likely to reveal defects in the App under test. We prioritized test paths for Mobile Apps based on input complexity, since more inputs might be associated with a more complex functionality which in turn would make it more fault-prone.

As we knew, regression testing is an important activity in software maintenance and enhancement. Combining several regression testing techniques can lead to a more efficient and effective regression test suite. In this dissertation, we presented guidelines for combining regression testing approaches based on a systematic approach. We outlined all possible situations that can occur and showed how each of them influences which combination to use.

Also, we validated the newly proposed regression testing approaches for Mobile Apps and the guidelines for combining regression testing approaches via a case study. The results show that FSMApp approaches are applicable, efficient, and effective.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Zeinab Saad Abdalla

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

275 pgs

Discipline

Computer science



Share

COinS