Inlämning av Examensarbete / Submission of Thesis

Shahid Mahmood MSE-2007:26, pp. 57. TEK/avd. för programvaruteknik, 2007.

The work

Författare / Author: Shahid Mahmood
shahidmahmood07@yahoo.com
Titel / Title: A Systematic Review of Automated Test Data Generation Techniques
Översatt titel / Translated title: A Systematic Review of Automated Test Data Generation Techniques
Abstrakt Abstract:

Automated Test Data Generation (ATDG) is an activity that in the course of software
testing automatically generates test data for the software under test (SUT). It usually makes
the testing more efficient and cost effective. Test Data Generation (TDG) is crucial for
software testing because test data is one of the key factors for determining the quality of any
software test during its execution. The multi-phased activity of ATDG involves various
techniques for each of its phases. This research field is not new by any means, albeit lately
new techniques have been devised and a gradual increase in the level of maturity has brought
some diversified trends into it. To this end several ATDG techniques are available, but
emerging trends in computing have raised the necessity to summarize and assess the current
status of this area particularly for practitioners, future researchers and students. Further,
analysis of the ATDG techniques becomes even more important when Miller et al. [4]
highlight the hardship in general acceptance of these techniques. Under this scenario only a
systematic review can address the issues because systematic reviews provide evaluation and
interpretation of all available research relevant to a particular research question, topic area,
or phenomenon of interest.
This thesis, by using a trustworthy, rigorous, and auditable methodology, provides a
systematic review that is aimed at presenting a fair evaluation of research concerning ATDG
techniques of the period 1997-2006. Moreover it also aims at identifying probable gaps in
research about ATDG techniques of defined period so as to suggest the scope for further
research. This systematic review is basically presented on the pattern of [5 and 8] and
follows the techniques suggested by [1].The articles published in journals and conference
proceedings during the defined period are of concern in this review. The motive behind this
selection is quite logical in the sense that the techniques that are discussed in literature of this
period might reflect their suitability for the prevailing software environment of today and are
believed to fulfill the needs of foreseeable future. Furthermore only automated and/or semiautomated
ATDG techniques have been chosen for consideration while leaving the manual
techniques as they are out of the scope.
As a result of the preliminary study the review identifies ATDG techniques and
relevant articles of the defined period whereas the detailed study evaluates and interprets all
available research relevant to ATDG techniques. For interpretation and elaboration of the
discovered ATDG techniques a novel approach called ‘Natural Clustering’ is introduced.
To accomplish the task of systematic review a comprehensive research method has been
developed. Then on the practical implications of this research method important results have
been gained. These results have been presented in statistical/numeric, diagrammatic, and
descriptive forms. Additionally the thesis also introduces various criterions for classification
of the discovered ATDG techniques and presents a comprehensive analysis of the results of
these techniques. Some interesting facts have also been highlighted during the course of
discussion. Finally, the discussion culminates with inferences and recommendations which
emanate from this analysis. As the research work produced in the thesis is based on a rich
amount of trustworthy information, therefore, it could also serve the purpose of being an upto-
date guide about ATDG techniques.

Ämnesord / Subject: Datavetenskap - Computer Science\Software Engineering
Datavetenskap - Computer Science\Computersystems
Datavetenskap - Computer Science\General
Nyckelord / Keywords: Systematic review, Automated test data generation techniques

Publication info

Dokument id / Document id:
Program:/ Programme Programvaruteknik/Software Engineering
Registreringsdatum / Date of registration: 11/27/2007
Uppsatstyp / Type of thesis: D-Uppsats/Magister/Master

Context

Handledare / Supervisor: Dr. Richard Torkar
richard.torkar@bth.se
Examinator / Examiner: Dr. Robert Feldt
Organisation / Organisation: Blekinge Institute of Technology
Institution / School: TEK/avd. för programvaruteknik
S-372 25 Ronneby
+46 455 38 50 00
Anmärkningar / Comments:

Shahid Mahmood
Folkparksvägen 14:23
372 40 Ronneby
Sweden
+46 76 2971676

Files & Access

Bifogad uppsats fil(er) / Files attached: shahidmtse2007-26.pdf (352 kB, öppnas i nytt fönster)