A Data Science Central Community
Many customer behaviors have the flavor of a choice between two alternatives: Yes or no. Buy or sell. Renew or cancel. Suppose software called a “classifier” is available to predict customer choices in advance. Would you use it? Perhaps you’d like to test it to see how well it performs before you commit. In this installment of my series on the nuts and bolts of data mining, I discuss the use of classifiers and questions about their performance. Regarding performance, we specifically consider hits, misses, false alarms, and the ROC Curve that pulls them all together.