Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


kent1271679781.pdf (13.27 MB)

ETD Abstract Container

Abstract Header

Empirical Assessment of UML Class Diagram Layouts Based on Architectural Importance

Sharif, Bonita
http://rave.ohiolink.edu/etdc/view?acc_num=kent1271679781

Abstract Details

2010, PHD, Kent State University, College of Arts and Sciences / Department of Computer Science.
The dissertation presents a suite of empirical studies that investigate the effectiveness of different layout techniques for Unified Modeling Language (UML) class diagrams. Three different layout schemes are examined. The architectural importance of classes is used to guide each layout technique. Architectural importance in a UML class diagram is defined by stereotype information, in this case the control, boundary and entity class stereotypes. Stereotypes are an extension mechanism provided by the UML that allow users to define semantics for the notation. The premise is that layout techniques for UML class diagrams significantly impact comprehension.The experiments use traditional questionnaire-based methods as well as eye-tracking equipment to quantitatively measure the performance of subjects solving specific software maintenance tasks. Both high-level and low-level software engineering tasks are examined. The layout techniques are also applied to design pattern comprehension in class diagrams. Two of the experiments are replicated using an alternate method of data collection to support and verify the findings. In addition, the effect of identifier styles (camel case and underscore) on the readability of UML class diagrams are also examined and is viewed as complementary to layout schemes used. The main contribution is a detailed empirical validation of a set of layout techniques with respect to a variety of software maintenance tasks. Results indicate that layout plays a significant role in the comprehension of UML class diagrams. In particular, there is a significant improvement in accuracy, time, and visual effort for one particular layout scheme, namely multi-cluster. In addition, a set of quantifiable eye-tracking measures are presented that provide an objective metric to measure the visual effort for class diagram layouts. The ultimate purpose of the research is to determine effective ways to adjust the layout of existing UML class diagrams to support program comprehension. Based on the empirical evidence, a set of guidelines for layout adjustment is presented.
Jonathan Maletic, PhD (Advisor)
L. Gwenn Volkert, PhD (Committee Member)
Robert Walker, PhD (Committee Member)
Michael Collard, PhD (Committee Member)
Jocelyn Folk, PhD (Committee Member)
419 p.
Computer Science; Education; Experiments
UML class diagram layout; controlled experiments; eye-tracking studies; empirical studies; layout guidelines; eye-tracking metrics; quality of UML class diagram layouts

Recommended Citations

Citations

  • Sharif, B. (2010). Empirical Assessment of UML Class Diagram Layouts Based on Architectural Importance [Doctoral dissertation, Kent State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=kent1271679781

    APA Style (7th edition)

  • Sharif, Bonita. Empirical Assessment of UML Class Diagram Layouts Based on Architectural Importance. 2010. Kent State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=kent1271679781.

    MLA Style (8th edition)

  • Sharif, Bonita. "Empirical Assessment of UML Class Diagram Layouts Based on Architectural Importance." Doctoral dissertation, Kent State University, 2010. http://rave.ohiolink.edu/etdc/view?acc_num=kent1271679781

    Chicago Manual of Style (17th edition)

kent1271679781
1,907
© 2010, all rights reserved.
This open access ETD is published by Kent State University and OhioLINK.