A Comparison of Constructive Induction withDi erent Types of

This paper studies the eeects on decision tree learning of constructing four types of attribute (conjunctive, disjunctive, M-of-N, and X-of-N representations). To reduce eeects of other factors such as tree learning methods, new attribute search strategies, search starting points, evaluation functions, and stopping criteria, a single tree learning algorithm is developed. With diierent option settings, it can construct four diierent types of new attribute, but all other factors are xed. The study reveals that conjunctive and disjunctive representations have very similar performance in terms of prediction accuracy and theory complexity on a variety of concepts, even on DNF and CNF concepts that are usually thought to be suited only to one of the two kinds of representation. In addition, the study demonstrates that the stronger representation power of M-of-N than conjunction and disjunction and the stronger representation power of X-of-N than these three types of new attribute can be reeected in the performance of decision tree learning in terms of higher prediction accuracy and lower theory complexity.