A construction approach of model transformation rules based on rough set theory

Abstract

Model transformation rules are the central part of model transformation. Many model transformation approaches provide some mechanisms to construct transformation rules in industrial and academic research. However, transformation rules are typically created manually in these approaches. As far as we know, there are no complete solutions that construct transformation rules automatically. In this paper, we propose a rough set based approach to construct transformation rules semi-automatically. Construction approach of rough set is improved in order to support the transformations between different meta-models, then the corresponding algorithm to construct transformation rules is presented. We also provide the measurement indicators of transformation rules to support selecting proper rules from many rules which meet transformation requirement. Three kinds of experiments for problems with distinct complexity and size are given for the validation of the proposed method. © 2011 IFIP International Federation for Information Processing.