Learning Disjunction of Conjunctions

The question of whether concepts expressible as disjunctions of conjunctions can be learned from examples in polynomial time is investigated. Positive results are shown for significant subclasses that allow not only propositional predicates but also some relations. The algorithms are extended so as to be provably tolerant to a certain quantifiable error rate in the examples data. It is further shown that under certain restrictions on these subclasses the learning algorithms are well suited to implementation on neural networks of threshold elements. The possible importance of disjunctions of conjunctions as a knowledge representation stems from the observations that on the one hand humans appear to like using it andon the other, that there is circumstantial evidence that significantly larger classes may not be learnable in polynomial time. An NP-completeness result corroborating the latter is also presented.