Is rotation forest the best classifier for problems with continuous features?
This work provides a practical default algorithm recommendation for machine learning practitioners dealing with continuous feature datasets, though it is incremental as it builds on existing rotation forest methods.
The paper investigates whether rotation forest is the best classifier for problems with continuous features, finding that it is significantly more accurate on average than alternatives like support vector machines, tree-based ensembles, and neural networks, with improvements demonstrated across three datasets and metrics such as classification error, AUC, and negative log likelihood. It also addresses scalability issues by proposing a contract version that reduces training time by an order of magnitude without significant accuracy loss.
In short, our experiments suggest that yes, on average, rotation forest is better than the most common alternatives when all the attributes are real-valued. Rotation forest is a tree based ensemble that performs transforms on subsets of attributes prior to constructing each tree. We present an empirical comparison of classifiers for problems with only real-valued features. We evaluate classifiers from three families of algorithms: support vector machines; tree-based ensembles; and neural networks tuned with a large grid search. We compare classifiers on unseen data based on the quality of the decision rule (using classification error) the ability to rank cases (area under the receiver operating characteristic) and the probability estimates (using negative log likelihood). We conclude that, in answer to the question posed in the title, yes, rotation forest is significantly more accurate on average than competing techniques when compared on three distinct sets of datasets. Further, we assess the impact of the design features of rotation forest through an ablative study that transforms random forest into rotation forest. We identify the major limitation of rotation forest as its scalability, particularly in number of attributes. To overcome this problem we develop a model to predict the train time of the algorithm and hence propose a contract version of rotation forest where a run time cap is imposed {\em a priori}. We demonstrate that on large problems rotation forest can be made an order of magnitude faster without significant loss of accuracy. We also show that there is no real benefit (on average) from tuning rotation forest. We maintain that without any domain knowledge to indicate an algorithm preference, rotation forest should be the default algorithm of choice for problems with continuous attributes.