Classification Performance Evaluation
A great part of this book presented the fundamentals of the classification process, a crucial field in data mining. It is now the time to deal with certain aspects of the way in which we can evaluate the performance of different classification (and decision) models. The problem of comparing classifiers is not at all an easy task. There is no single classifier that works best on all given problems, phenomenon related to the ”No-free-lunch” metaphor, i.e., each classifier (’restaurant’) provides a specific technique associated with the corresponding costs (’menu’ and ’price’ for it). It is hence up to us, using the information and knowledge at hand, to find the optimal trade-off.
KeywordsPositive Predictive Value Negative Predictive Value False Negative Rate Testing Dataset Confusion Matrix
Unable to display preview. Download preview PDF.