Open AccessUML models change impact analysis using a text similarity technique
Author(s) -
Kchaou Dhikra,
Bouassida Nadia,
Ben-Abdallah Hanêne
Publication year - 2017
Publication title -
iet software
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.305
H-Index - 43
eISSN - 1751-8814
pISSN - 1751-8806
DOI - 10.1049/iet-sen.2015.0113
Subject(s) - computer science , data mining , unified modeling language , search engine indexing , traceability , sequence diagram , change impact analysis , context (archaeology) , information retrieval , software , software engineering , programming language , paleontology , biology
Given the inevitable software evolution, change impact analysis (CIA) is a vital activity in the software development life cycle. Existing CIA methods either focus on one model produced during one development phase or ignore the semantic dependencies among the various models produced throughout the development phases. The herein proposed CIA method overcomes this limit by exploiting the structural and semantic dependencies within and inter‐UML diagrams. It uses a graph technique to model the structural dependencies and an information retrieval (IR) technique to handle the semantic traceability between the use case documentation and the sequence diagrams. To identify the most appropriate IR technique to the CIA context, the authors present a quantitative experimentation of term frequency‐inverse document frequency and latent semantic indexing (LSI), two widely used IR techniques. In addition, to evaluate the overall performance of their method, they present an evaluation performed on changes in the open source system JHotDraw 7.4.1 compared with its 7.5.1 version and changes performed on a real existing application. Using LSI, their method achieves an average precision of 84% and a recall of 91% in the requirements CIA and management.