From: Flight delay prediction based on deep learning and Levenberg-Marquart algorithm
 | Predicted delay | Predicted non-delay |
---|---|---|
Actual Delay | True Positive (TP) = number of behaviors that represent delay and is forecasted by model | False Positive (FN) = number of behaviors that represent delay and the model has incorrectly predicted no delay |
Actual Non-Delay | False Negative (FP) = Number of behaviors that show no delay and model incorrectly predicted as a delay | True Negative (TN) = Number of behaviors that show no delay and is forecasted by model |