Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Automatic Requirement Classification Technique: Using Different Stemming Algorithms


     

   Subscribe/Renew Journal


Requirement Engineering is the first crucial stage in the software life cycle. Classifying those requirements into functional and non-functional requirements is an important activity during requirement engineering process. As a result of requirement engineering process, a software requirement specification document is produced. This document contains a detailed description of all requirements written using natural language. The automatic processing of natural language is not an easy task. Since natural language is full of ambiguity, has no formal structure, and very variable. This paper presents an automatic classification of requirements into functional and non-functional requirements using two machine learning algorithms. In this paper, different stemming techniques are used to address some of natural language challenges. A dataset of 625 requirements (functional and non-functional) is used to train and test the machine learning model. The experiments showed that some stemming techniques increased the performance than other stemming techniques.

 


Keywords

Requirement Classification, Non-Functional Requirements, Stemming, Software Projects, Functional Requirements.
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 254

PDF Views: 0




  • Automatic Requirement Classification Technique: Using Different Stemming Algorithms

Abstract Views: 254  |  PDF Views: 0

Authors

Abstract


Requirement Engineering is the first crucial stage in the software life cycle. Classifying those requirements into functional and non-functional requirements is an important activity during requirement engineering process. As a result of requirement engineering process, a software requirement specification document is produced. This document contains a detailed description of all requirements written using natural language. The automatic processing of natural language is not an easy task. Since natural language is full of ambiguity, has no formal structure, and very variable. This paper presents an automatic classification of requirements into functional and non-functional requirements using two machine learning algorithms. In this paper, different stemming techniques are used to address some of natural language challenges. A dataset of 625 requirements (functional and non-functional) is used to train and test the machine learning model. The experiments showed that some stemming techniques increased the performance than other stemming techniques.

 


Keywords


Requirement Classification, Non-Functional Requirements, Stemming, Software Projects, Functional Requirements.