BDBComp
Parceria:
SBC
Combining Aspect Mining Techniques Based on Crosscutting Concern Sorts

Esteban S. AbaitClaudia A. Marcos

Aspect mining encompasses techniques for identifying crosscutting concern code in legacy systems. Those techniques share several problems, including low recall and precision. In this paper, a dynamic aspect mining technique based on crosscutting concern sorts is introduced, and combined with four previous static techniques in two case studies. By combining our dynamic based technique with the other static ones we were able to improve the percentage of identified candidates 72% and 178% on average for each analyzed application. Based on the results, we reckon that static and dynamic based approaches yield complementary results and hence should be used in conjunction.

http://www.lbd.dcc.ufmg.br:8080/colecoes/lawasp/2009/004.pdf

Caso o link acima esteja inválido, faça uma busca pelo texto completo na Web: Buscar na Web

Biblioteca Digital Brasileira de Computação - Contato: bdbcomp@lbd.dcc.ufmg.br
     Mantida por:
LBD