Unsupervised Learning for Detecting Refactoring Opportunities in Service-Oriented Applications

RODRIGUEZ, GUILLERMO; SORIA, ALVARO; TEYSEYRE, ALFREDO; BERDUN, LUIS; CAMPO, MARCELO

Porto

Congreso; Database and Expert Systems Applications; 2016

Service-Oriented Computing (SOC) has been widelyused for building distributed and enterprise-wide software applications.One major problem in this kind of applications istheir growth; as size and complexity of applications increase,the probability of duplicity of code increases, among otherrefactoring issues. This paper proposes an unsupervised learningapproach to assist software developers in detecting refactoringopportunities in service-oriented applications. The approachgathers non-refactored Web Service Description Language(WSDL) documents and applies clustering and visualizationtechniques to deliver a list of refactoring suggestions tostart working on the refactoring process. We evaluated our approachusing two real-life case-studies by using internal validitycriteria for the clustering quality.