Table 6 Evaluation criteria for classification algorithm.

From: Optimized extreme learning machines with deep learning for high-performance network traffic classification

Evaluation Metrics

Equation

Description

TP

-

True Positive

TN

-

True Negative

FP

-

False Positive

FN

-

False Negative

Accuracy

(5)

The percentage of correctly classified instances.

Sensitivity (TPR)

(6)

True Positive Rate

Specificity (TNR)

(7)

True Negative Rate

Precision

(8)

The degree of validity for positive responses.

Recall

(9)

The classifier’s ability to detect unsafe traffic.

F1-Measure

(10)

A combination of precision and recall.

Training Time

-

The time it takes to train the classifier.

Prediction Time

-

The time it takes to classify a single incoming traffic.