ElicitO: A Quality Ontology-Guided NFR Elicitation Tool

Abstract

Despite the importance of capturing a precise and complete set of requirements in the requirements engineering stage, there are few tools that adequately support requirements analysis in the process of capturing quality related requirements (non-functional requirements). This paper presents ElicitO, a requirements elicitation tool aimed at empowering requirements analysts with a knowledge repository that helps in the process of capturing precise nonfunctional requirements (NFRs) specifications during elicitation interviews. The approach is based on the application of functional and non-functional domain ontologies (quality ontologies) to underpin the elicitation activities. The tool is used as a memory aid to structure elicitation interviews, guide requirements analysts with regard to the important quality aspects relating to a class of applications, and support the development of precise requirements based on characteristics and metrics available in quality model standards.

Identification of non-functional requirements in textual specifications: A semi-supervised learning approach

“…Further classification of the elicited requirements could be assisted by domain-specific knowledge (i.e. the Language Extended Lexicon or LEL, which uses the vocabulary of the domain together with a NFR knowledge base [10]) or by ontologies representing quality aspects that should be taken into consideration [2,20].…”

Classification and Qualitative Analysis of Non-Functional Requirements Approaches

“…Goal-oriented Aspect-oriented Pattern-based • Use-case Questioner [14] • MOQARE [15] • NFRs elicitation model [16] • Actor-based model [17] • Usability catalogue [18] • NFRs Layered Framework [19] • Goal-based requirement extraction [20] • Usability Elicitation Framework [21] • NFR Classifier [22] • Semi supervised text Analysis [23] • QA-Miner [24] • NFR Incorporation Framework [25] • Speech recognition [26] • Experience-based Method [27] • NFRs elicitation Framework [28] • ElicitO [29] • Efficiency usecases [30] • NFR recommendation [31] In goal-oriented elicitation approach, goal-based questionnaire is proposed to extract NFRs by questioning stakeholders [14,15], business process model [16], usecase of domain model [17], and taxonomy [18] in the system development. Goal decomposing [19] and goal analysis [20] methods are proposed to identify stakeholder, generate their expected goals based on developer's knowledge and experience, then decompose the goals into sub-goals and identify NFRs for each sub-goals.…”

“…meeting minutes, interview notes, and memos). In pattern-based approach, experience-based elicitation [27], domain ontology [28][29][30] are proposed to assist requirements analysts in NFRs elicitation process. Recommendation system [31] is also proposed to understand expected realistic NFRs in the system development.…”

Visualization and comparison of architecture rationale with semantic web technologies

“…Dobson et al (2007) presented a domain-independent ontology for NFRs, which models concepts and relationships related to quality attributes and their metrics. Hazeem proposed ElicitO (Balushi et al, 2007), an ontology-based tool that supports NFR elicitation and provides a knowledge base of NFR requirements and related metrics. Neither Donson's ontology nor ElicitO describe NFR solutions, and therefore do not facilitate recording and manipulating architecture rationale.…”

“…This will allow improving and augmenting existing NFR and architectural knowledge in the Web. Also, relating the NDR Ontology and some metric-related ontologies (Balushi et al, 2007;Dobson et al, 2007) would augment the evaluation process support in early phases. Finally, defining mappings between the NDR Ontology and architectural decision-related ontologies (Akerman and Tyree, 2006;Kruchten et al, 2006) will enable, among other issues, to integrate the based-view exploration and comparison capabilities to larger and more complex NFR knowledge bases.…”