DFG Project on Rule Learning extended