Concept analysis as a framework for mining functional features from legacy code

  • Authors:
  • Amal El Kharraz;Petko Valtchev;Hafedh Mili

  • Affiliations:
  • Dépt. d’Informatique UQAM, C.P. 8888, Succ. Centre-Ville, Montréal, Canada;Dépt. d’Informatique UQAM, C.P. 8888, Succ. Centre-Ville, Montréal, Canada;Dépt. d’Informatique UQAM, C.P. 8888, Succ. Centre-Ville, Montréal, Canada

  • Venue:
  • ICFCA'10 Proceedings of the 8th international conference on Formal Concept Analysis
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Legacy OO applications typically implement a set of functional features which, in the absence of aspect-oriented techniques to separately develop and maintain them, end up embodied in the same class hierarchies. We identified three types of design techniques used to implement that embodiment: a) multiple inheritance– or simulations thereof, b) aggregation/delegation, and c) what we referred to as ad-hoc implementation. We are interested in identifying and isolating software artifacts that implement distinct functional features. Here, we explore the use of concept analysis to detect ad-hoc implementations of functional features. We present the principles underlying our overall approach, a formalization of the problem in terms of concept analysis, a method for identifying functional features based on the construction and exploration of the concept latice, and the results of an experimental validation study.