Is your classification model making lucky guesses?

March 14, 2016
This experiment demonstrates how to evaluate Classification Model against some baseline metrics.
This experiment demonstrates how to evaluate Classification Model against some baseline metrics. This helps establish that the model is not just making lucky guesses. For a binary classification using census income data, an accuracy of 76% can be achieved simply by assigning all instances to the majority class! <img src="https://raw.githubusercontent.com/shaheeng/ClassificationModelEvaluation/master/Pictures/nonMLmodels3.png" width="300" height="400" />