An Event-Based Methodology to Generate Class Diagrams and its Empirical Evaluation
Sandeep K. Singh, Sangeeta Sabharwal & J. P. Gupta
Journal of Computer Science, Volume 6, Issue 11, pp. 1301–1325 · doi.org/10.3844/jcssp.2010.1301.1325
Abstract
Problem statement: Event-based systems have importance in many application domains ranging from real time monitoring systems in production, logistics, medical devices and networking to complex event processing in finance and security. The increasing popularity of Event-based systems has opened new challenging issues for them. One such issue is to carry out requirements analysis of event-based systems and build conceptual models. Currently, Object Oriented Analysis (OOA) using Unified Modeling Language (UML) is the most popular requirement analysis approach for which several OOA tools and techniques have been proposed. But none of the techniques and tools to the best of our knowledge, have focused on event-based requirements analysis, rather all are behavior-based approaches. Approach: This study described a requirement analysis approach specifically for event based systems. The proposed approach started from events occurring in the system and derives an importable class diagram specification in XML Metadata Interchange (XMI) format for Argo UML tool. Requirements of the problem domain are captured as events in restricted natural language using the proposed Event Templates in order to reduce the ambiguity. Results: Rules were designed to extract a domain model specification (analysis-level class diagram) from Event Templates. A prototype tool 'EV-ClassGEN' is also developed to provide automation support to extract events from requirements, document the extracted events in Event Templates and implement rules to derive specification for an analysis-level class diagram. The proposed approach is also validated through a controlled experiment by applying it on many cases from different application domains like real time systems, business applications, gaming. Conclusion: Results of the controlled experiment had shown that after studying and applying Event-based approach, student's perception about ease of use and usefulness of OOA technique has significantly improved. Their project reported showed positive feedback about Event-based approach. These results reinforced the evidence that by analyzing events that are likely to happen in a system, one can derive class diagram information from requirements.
Licence and reuse
Free to read. This article carries the publisher’s copyright, as printed on it: © 2010 Science Publications. It is free to read; ask us before reusing it.
How to reuse this articleResponses (0)
About this article
Singh, S. K., Sabharwal, S. & Gupta, J. P. (2010). An Event-Based Methodology to Generate Class Diagrams and its Empirical Evaluation. Journal of Computer Science, 6(11), 1301–1325. https://doi.org/10.3844/jcssp.2010.1301.1325
- Licence
- © Science Publications
- Issue
- Volume 6, Issue 11
- Policies
- Editorial independenceUse of generative AIData availabilityAppeals and complaints
- Version
- Version of record
