A comparison and evaluation of data requirement specification techniques in SSADM and the unified process

3Citations
Citations of this article
5Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

During the analysis stage of a typical information system development process, user requirements concerning system functionality and data are captured and specified using requirement specification techniques. Most of these specification techniques are graphical (semi-formal), i.e. they involve modelling. This paper presents a comparison of data requirement specification techniques in SSADM, which is a strong data-centred method, and the Unified Process, an object-oriented method. In particular, we investigated how data groups (entities and classes), their attributes and relationships are identified, specified and validated in both methods. Data requirement specification techniques used in both methods are then evaluated against a set of detailed criteria based on five requirement specification quality attributes. Both methods seem to have a similar informal approach to producing initial data requirement specifications, but they differ when these initial models are refined. The refinement in SSADM is more rigorous. Therefore, this paper makes a few recommendations for the Unified Process.

Cite

CITATION STYLE

APA

Bielkowicz, P., & Tun, T. T. (2001). A comparison and evaluation of data requirement specification techniques in SSADM and the unified process. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2068, pp. 46–59). Springer Verlag. https://doi.org/10.1007/3-540-45341-5_4

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free