Conversion of Decision Tables into Decision Trees

摘要

Known manual methods of converting decision tables into decision trees are based mainly on plausible arguments with little theoretical backgrounds. The intention of the author is to establish a new theory in this field. By considering a special kind of partitions of (2 sup n) vertices of an n-cube as a model of decision tables, the author puts the conversion problem into a simplified and abstract form. The author derives some theoretical results concerning the optimization problem, and then an algorithm called iterated local minimization is proposed and compared quantitatively with other algorithms. Also, the new topic, a decomposition theory of decision tables and decision trees is presented. The author considers decomposing decision tables on decision trees into smaller ones so that they can be processed effectively in parallel. (Author)