CRESUS-T: A COLLABORATIVE REQUIREMENTS
ELICITATION SUPPORT TOOL
Paul Stynes1
, Owen Conlan2
and Declan O’Sullivan3
1
School of Computing, National College or Ireland, Dublin, Ireland
2,3ADAPT centre, Trinity College Dublin, Dublin 2, Ireland.
ABSTRACT
Communicating an organisation's requirements in a semantically consistent and understandable manner
and then reflecting the potential impact of those requirements on the IT infrastructure presents a major
challenge among stakeholders. Initial research findings indicate a desire among business executives for a
tool that allows them to communicate organisational changes using natural language and a model of the IT
infrastructure that supports those changes. Building on a detailed analysis and evaluation of these findings,
the innovative CRESUS-T support tool was designed and implemented. The purpose of this research was to
investigate to what extent CRESUS-T both aids communication in the development of a shared
understanding and supports collaborative requirements elicitation to bring about organisational, and
associated IT infrastructural, change. In order to determine the extent shared understanding was fostered,
the support tool was evaluated in a case study of a business process for the roll out of the IT software
image at a third level educational institution. Statistical analysis showed that the CRESUS-T support tool
fostered shared understanding in the case study, through increased communication. Shared understanding
is also manifested in the creation of two knowledge representation artefacts namely, a requirements model
and the IT infrastructure model. The CRESUS-T support tool will be useful to requirements engineers and
business analysts that have to gather requirements asynchronously.
KEYWORDS
Collaborative Requirements Elicitation, Shared Understanding, and Semantically enabled Web Services,
Ontology.
Original Source URL: http://aircconline.com/ijsea/V9N2/9218ijsea02.pdf
For More Details: http://www.airccse.org/journal/ijsea/ijsea.html
Comments
Post a Comment