Detection of attacks based on compromise marks | IEEE Conference Publication | IEEE Xplore