Abstract
Software quality measurement lies at the heart of the quality engineering process. Quality measurement for object-oriented artifacts has become the key for ensuring high quality software. Both researchers and practitioners are interested in measuring software product quality for improvement. It has recently become more important to consider the quality of products at the early phases, especially at the design level to ensure that the coding and testing would be conducted more quickly and accurately. The research work on measuring quality at the design level progressed in a number of steps. The first step was to discover the correct set of metrics to measure design elements at the design level. Chidamber and Kemerer (C&K) formulated the first suite of OO metrics. Other researchers extended on this suite and provided additional metrics. The next step was to collect these metrics by using software tools. A number of tools were developed to measure the different suites of metrics; some represent their measurements in the form of ordinary numbers, others represent them in 3D visual form. In recent years, researchers developed software quality models which went a bit further by computing quality attributes from collected design metrics. In this research we extended on the software quality modelers’ work by adding a quality attribute prioritization scheme and a design metric analysis layer. Our work is all focused on the class diagram, the most fundamental constituent in any object oriented design. Using earlier researchers’ work, we extract a class diagram’s metrics and compute its quality attributes. We then analyze the results and inform the user. We present our figures and observations in the form of an analysis report. Our target user could be a project manager or a software quality engineer or a developer who needs to improve the class diagram’s quality. We closely examine the design metrics that affect quality attributes. We pinpoint the weaknesses in the class diagram, based on these metrics, inform the user about the problems that emerged from these classes, and advice him/her as to how he/she can go about improving the overall design quality. We consider the six basic quality attributes: “Reusability”, “Functionality”, “Understandability”, “Flexibility”, “Extendibility”, and “Effectiveness” of the whole class diagram. We allow the user to set priorities on these quality attributes in a sequential manner based on his/her requirements. Using a geometric series, we calculate a weighted average value for the arranged list of quality attributes. This weighted average value indicates the overall quality of the product, the class diagram. Our experimental work gave us much insight into the meanings and dependencies between design metrics and quality attributes. This helped us refine our analysis technique and give more concrete observations to the user.
School
School of Sciences and Engineering
Department
Computer Science & Engineering Department
Degree Name
MS in Computer Science
Date of Award
6-1-2009
Online Submission Date
3-10-2013
First Advisor
Hosny, Hoda
Committee Member 1
Hosny, Hoda
Document Type
Thesis
Extent
159 p.
Rights
The American University in Cairo grants authors of theses and dissertations a maximum embargo period of two years from the date of submission, upon request. After the embargo elapses, these documents are made available publicly. If you are the author of this thesis or dissertation, and would like to request an exceptional extension of the embargo period, please write to thesisadmin@aucegypt.edu
IRB
Not necessary for this item
Recommended Citation
APA Citation
Rizk, D.
(2009).Software quality attribute measurement and analysis based on class diagram metrics [Thesis, the American University in Cairo]. AUC Knowledge Fountain.
https://fount.aucegypt.edu/retro_etds/2321
MLA Citation
Rizk, Dalia. Software quality attribute measurement and analysis based on class diagram metrics. 2009. American University in Cairo, Thesis. AUC Knowledge Fountain.
https://fount.aucegypt.edu/retro_etds/2321