Studies of inference rule creation using LAPART

The logical neural architecture LAPART is used in a mode that allows through learning the easy creation and extraction of IF-THEN inference rules from data. This paper first describes ART1 and the complement coded stack input binary representations. Next, we present a more detailed discussion of LAPART. Then we show how rules are learned and extracted from the memory templates of the ART1s. We present a pedagogical example of rules extracted from a simple data set. Finally, we note that a fundamental difference between LAPART rule-based systems and regular rule-based systems is the existence of a “rule attractor” that can enhance system generalization in a controlled manner