Detection and Analysis of Ambiguities in Software Requirement Specification

Main Article Content

Prerana Chaithra, Dr. Shantharam Nayak

Abstract

Software requirement specification (SRS) forms an important document for capturing the details of newly designed software. Since natural language is used for representing, this enables the end-user to assess the suitability of the software for the application. Hence, the quality of SRS document is of great importance and evaluation of the document plays a significant role. The present study aims at identifying the lexical and syntactic ambiguities in the SRS document. Identification of ambiguities is carried out by tagging the extracted words from the document assisted by natural language processing. Python with in-built Natural Language Tool Kit (NLTK) is used for extracting the parts-of-speech (POS), while storing and retrieving the data is carried out with the aid of MongoDB database. The ambiguity levels are defined using the Ambiguity datahandbook[1] and the document ambiguity is assessed. The overall ambiguity percentage is specified and the changes necessary for the improvement of the document are indicated.


 


Keywords: Software Requirement Specification(SRS), Lexical Ambiguity(LE), Syntactic Ambiguity(SE), Natural Language Processing (NLP), Parts-of-speech (POS), stop-words, tokenization

Article Details

Section
Articles