Pattern-based decision tree construction

Learning classifiers has been studied extensively the last two decades. Recently, various approaches based on patterns (e.g.. association rules) that hold within labeled data have been considered. In this paper, we propose a novel associative classification algorithm that combines rules and a decision tree structure. In a so-called delta-PDT (delta-pattern decision tree), nodes are made of selected disjunctive delta- strong classification rules. Such rules are generated from collections of delta-free patterns that can be computed efficiently. These rules have a minimal body, they are non- redundant and they avoid classification conflicts under a sensible condition on delta. We show that they also capture the discriminative power of emerging patterns. Our approach is empirically evaluated by means of a comparison to state-of-the-art proposals (i.e., C4.5, CBA CPAR, SJEPs- classifier).