MADALINE RULE II: a training algorithm for neural networks | IEEE Conference Publication | IEEE Xplore