Machine learning methods for classification problems commonly assume that the class values are unordered. However, in many practical applications the class values do exhibit a natural order -- for example, when learning how to grade. The standard approach to ordinal classification converts the class value into a numeric quantity and applies a regression learner to the transformed data, translating the output back into a discrete class value in a post-processing step. A disadvantage of this method is that it can only be applied in conjunction with a regression scheme. In this paper we present a simple method that enables standard classification algorithms to make use of ordering information in class attributes. By applying it in conjunction with a decision tree learner we show that it outperforms the naive approach, which treats the class values as an unordered set. Compared to special-purpose algorithms for ordinal classification our method has the advantage that it can be applied without any modification to the underlying learning scheme.
Ordinal classification is a form of multi-class classification where there is an inherent ordering between the classes, but not a meaningful numeric difference between them. Little attention has been paid as to how to evaluate these problems, with many authors simply reporting accuracy, which does not account for the severity of the error. Several evaluation metrics are compared across a dataset for a problem of classifying user reviews, where the data is highly skewed towards the highest values. Mean squared error is found to be the best metric when we prefer more (smaller) errors overall to reduce the number of large errors, while mean absolute error is also a good metric if we instead prefer fewer errors overall with more tolerance for large errors.