An Experimental Comparison of Old and New Decision Tree Algorithms
This work provides an experimental evaluation for machine learning practitioners, but it is incremental as it focuses on comparing a new algorithm against existing ones without introducing a fundamental change.
The paper compared the new tree alternating optimization (TAO) algorithm with established decision tree methods on various datasets, finding that TAO achieved higher accuracy in nearly all cases, often by a large margin.
This paper presents a detailed comparison of a recently proposed algorithm for optimizing decision trees, tree alternating optimization (TAO), with other popular, established algorithms. We compare their performance on a number of classification and regression datasets of various complexity, different size and dimensionality, across different performance factors: accuracy and tree size (in terms of the number of leaves or the depth of the tree). We find that TAO achieves higher accuracy in nearly all datasets, often by a large margin.