Science and Technology Production
Article
Authorship
Crasso, Marco Patricio
;
MATEOS DIAZ, CRISTIAN MAXIMILIANO
;
ZUNINO SUAREZ, ALEJANDRO OCTAVIO
;
Campo, Marcelo Ricardo
Date
2014
Publishing House and Editing Place
Elsevier Science Inc
Magazine
INFORMATION SCIENCES,
vol. 259
(pp. 452-473)
Elsevier Science Inc
Summary
Information provided by the agent in
SIGEVA
Service-Oriented Computing has been widely recognized as a revolutionary paradigm for software development. Despite the important benefits this paradigm provides, current approaches for service-enabling applications still lead to high costs for outsourcing services with regard to two phases of the software life cycle. During the implementation phase, developers have to invest much effort into manually discovering services and then providing code to invoke them. Mostly, the outcome of the second...
Service-Oriented Computing has been widely recognized as a revolutionary paradigm for software development. Despite the important benefits this paradigm provides, current approaches for service-enabling applications still lead to high costs for outsourcing services with regard to two phases of the software life cycle. During the implementation phase, developers have to invest much effort into manually discovering services and then providing code to invoke them. Mostly, the outcome of the second task is software containing service-aware code, therefore it is more difficult to modify and to test during the maintenance phase. This paper describes EasySOC, an approach that aims to decrease the costs of creating and maintaining service-oriented applications. EasySOC combines text mining, machine learning, and best practices from component-based software development to allow developers to quickly discover and non-invasively invoke services. We evaluated the performance of the EasySOC discovery mechanism using 391 services. In addition, through a case study, we conducted a comparative analysis of the software technical quality achieved by employing EasySOC versus not using it.
Show more
Show less
Key Words
MACHINE LEARNINGTEXT MININGSERVICE OUTSOURCINGSERVICE-ORIENTED COMPUTINGDEPENDENCY INJECTION
Download or request the full text