Authors:
Akanksha Mishra
1
and
Richa Sharma
2
Affiliations:
1
Indraprastha Institute of Information Technology, India
;
2
BML Munjal University, India
Keyword(s):
Requirements Engineering, Incompleteness, Quality Attributes, Requirements, Case Frames, FrameNet.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Knowledge Management and Information Sharing
;
Knowledge-Based Systems
;
Requirements Engineering
;
Symbolic Systems
Abstract:
Requirements gathered during early phase of requirements engineering are informal and vague. These informal requirements are analyzed with the goal of detecting three major problems in requirements – ambiguity, inconsistency, and incompleteness in order to arrive at correct and formal set of requirements. These problems are quite intertwined, with one problem leading to another. Incompleteness in requirements, however, is considered to be a principal reason for poor quality of requirements, and is the most difficult issue to address. There are multiple views around defining and detecting incompleteness in requirements. In this paper, we present an approach towards handling incompleteness in informal requirements considering individual requirements statement expressed in natural language as an atomic requirement. Our approach is based on enriching frame-based structured representation using FrameNet database that, in turn, can prove useful in identifying potential missing information
from requirements. We also report our observations from the evaluation study conducted with a case study.
(More)