Learning Transformation Rules from Transformation Examples: An Approach Based on Relational Concept Analysis

In Model Driven Engineering (MDE), model transformations are basic and primordial entities, thus easing their design and implementation is an important issue. A quite recently proposed way to create model transformations consists in deducing a transformation from examples of transformed models. Examples are easier to write than a transformation program and are often already available. We propose in this paper a method based on a machine learning method of the lattice domain, the Relational Concept Analysis, and an implementation of this method.