Reference Work Entry

Encyclopedia of Machine Learning

pp 683-683

Model Evaluation

  • Geoffrey I. Webb

Model evaluation is the process of assessing a property or properties of a model.

Motivation and Background

It is often valuable to assess the efficacy of a model that has been learned. Such assessment is frequently relative—an evaluation of which of several alternative models is best suited to a specific application.

Processes and Techniques

There are many metrics by which a model may be assessed. The relative importance of each metric varies from application to application.

The primary considerations often relate to predictive efficacy—how useful will the predictions be in the particular context it is to be deployed. Measures relating to predictive efficacy include Accuracy, Lift, Mean Absolute Error, Mean Squared Error, Negative Predictive Value, Positive Predictive Value, Precision, Recall, Sensitivity, Specificity, and various metrics based on ROC analysis.

Computational issues may also be important, such as a model’s size or its executio ...

This is an excerpt from the content