Clone Detection for Ecore Metamodels using N-grams
Önder Babur · 2018
Increasing model-driven engineering use leads to an abundance of models and metamodels in academic and industrial practice. A key technique for the management and maintenance of those artefacts is model clone detection, where highly similar (meta-)models and (meta-)model fragments are mined from a possibly large amount of data. In this paper we extend the SAMOS framework (Statistical Analysis of MOdelS) to clone detection on Ecore metamodels, using the framework’s n-gram feature extraction, vector space model and clustering capabilities. We perform a case analysis on Ecore metamodels obtained by applying an exhaustive set of single mutations to assess the precision/sensitivity of our technique with respect to various types of mutations. Using mutation analysis, we also briefly evaluate MACH, a comparable UML clone detection tool.