Evaluating a meta-model for interoperability problem classification