Document Details

Document Type : Thesis 
Document Title :
Integrating Software Requirements Ambiguity Resolution Techniques with Empirical Evaluation
الغموض في متطلبات البرمجيات: تجنبه والكشف عنه مع تقييم عملي
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : High-level goals of a software project are initially elicited from the stakeholders in the form of natural language user requirements. Unfortunately, natural languages suffer from inherent ambiguity, where an expression may have more than one meaning or could be understood in different ways. This can lead to critical differences between the intended system and the actual one. Resolving ambiguity in the user requirements is much cheaper than dealing with errors in later stages of software development. Relatively few research studies in the literature addressed ambiguity resolution and they utilize various techniques for ambiguity prevention, avoidance, and detection. Unfortunately, none has been successful enough to be widely adopted. This also stems from the fact that there is a shortage in empirical evaluations of such techniques. Additionally, techniques for ambiguity avoidance are provided merely as general practices. Thus, this research developed the Ambiguity Avoidance Detection Prevention AADP tool, which implements various requirements ambiguity resolution techniques. This tool composed of three modules: 1) avoidance module that enforces ambiguity avoidance rules; 2) prevention module that fully implements an existing partially implemented work; 3) detection module that detects ambiguities consulting four resources such as ambiguity database. We gathered the most prevalent ambiguity types. Then we built a benchmark of 2462 real requirements. Additionally, we collected 282 ambiguous words and phrases to fill the ambiguity database to conduct an empirical qualitative and quantitative evaluation. The evaluation compares the three techniques in term of ambiguity types coverage and time performance. This study concluded that the two techniques ambiguity avoidance and ambiguity detection (using ambiguous phrases database) were the best candidates for integration into a final ambiguity resolution tool. That is because they cover all ambiguity types considered in this study. 
Supervisor : Prof. Hanan ElAzhary 
Thesis Type : Master Thesis 
Publishing Year : 1441 AH
2019 AD 
Added Date : Wednesday, September 25, 2019 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
رشا محمد العمريAlOmari, Rasha MohammedResearcherMaster 

Files

File NameTypeDescription
 45050.pdf pdf 

Back To Researches Page