Programming Drills with a Decision Trees Workbench

Abstract

Decision trees are one of the most successful Machine Learning paradigms. This paper presents a library of decision tree algorithms in Java that was eventually used as a programming laboratory workbench. The initial design focus was, as regards the non-expert user, to conduct experiments with decision trees using components and visual tools that facilitate tree construction and manipulation and as regards the expert user, to be able to focus on algorithm design and comparison with few implementation details. The system has been built over a number of years and over various development contexts and has been successfully used as a workbench in a programming laboratory for junior computer science students. The underlying philosophy was to achieve a solid introduction to object-oriented concepts and practices based on a fundamental machine learning paradigm.