DMCA
Aspect Mining using a Vector-Space Model Based Clustering Approach
| Venue: | In: Proceedings of Linking Aspect Technology and Evolution (LATE) Workshop, 2006 |
| Citations: | 9 - 2 self |
BibTeX
@INPROCEEDINGS{Moldovan_aspectmining,
author = {Grigoreta Sofia Moldovan},
title = {Aspect Mining using a Vector-Space Model Based Clustering Approach},
booktitle = {In: Proceedings of Linking Aspect Technology and Evolution (LATE) Workshop, 2006},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Aspect Oriented Programming is a programming paradigm that addresses the issue of crosscutting concerns. Aspect mining is a process that tries to identify crosscutting concerns in existing software systems. The goal is to refactor the existing systems to use aspect oriented programming to make them easier to maintain and to evolve. This paper presents a new approach in aspect mining that uses clustering and proposes two techniques: a k-means based clustering technique and a hierarchical agglomerative based clustering technique. We are trying to identify the methods that have the code scattering symptom. For a method, we consider as indication of code scattering a big number of calling methods and, also, a big number of calling classes. In order to group the best methods (candidates), we use in our approach the vector-space model for defining the similarity between methods. For testing the efficiency of the proposed techniques, a number of Java applications are being used.
