| Popis: |
The Service Oriented Computing paradigm and its most popular implementation, namely Web Services, are at the crossing of distributed computing and loosely coupled systems. Web Services can be discovered and reused dynamically using non-proprietary mechanisms, but when Web Services are poorly described, they become difficult to be discovered, understood, and then reused. This paper presents novel algorithms and heuristics for automatically detecting common pitfalls that should be avoided when creating Web Services descriptions. To assess the accuracy of the proposed algorithms and heuristics, we compared their results with the results of manually analyzing a data-set of 400 publicly available services. In addition, we analyzed the correlation between the algorithms and heuristics results and other well-known quality metrics, which were presented by Al-Masri and Mahmoud. The average detection accuracy was 93.14% , and the false positive and false negative rates of 4.06% and 9.91% , respectively. Additionally, the Al-Masri and Mahmoud´s quality metrics related to Web Services descriptions had a direct correlation with most of the automatic detecting results. The proposed algorithms and heuristics for automatically detecting common pitfalls are powerful tools for both improving existent Web Services and developing new Web Services that can be easily discovered, understood and reused. Fil: Rodriguez, Juan Manuel. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Tandil. Instituto Superior de Ingenieria del Software; Argentina Fil: Zunino Suarez, Alejandro Octavio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Tandil. Instituto Superior de Ingenieria del Software; Argentina Fil: Crasso, Marco Patricio. Consejo Nacional de Investigaciones Científicas y Técnicas. Centro Científico Tecnológico Tandil. Instituto Superior de Ingenieria del Software; Argentina |
