Elephant Flows Detection Using Deep Neural Network, Convolutional Neural Network, Long Short-Term Memory, and Autoencoder

Table 5

Best validation accuracy and threshold.

---

Best validation accuracy: 0.9795 for threshold: 0.1555

Perentile: 90, threshold: 0.1005, validation accuracy: 0.9521

Perentile: 91, threshold: 0.1028 validation, accuracy: 0.9543

Perentile: 92, threshold: 0.1083, validation accuracy: 0.9658

Perentile: 93, threshold: 0.1113, validation accuracy: 0.9658

Perentile: 94, threshold: 0.1173, validation accuracy: 0.9703

Perentile: 95, threshold: 0.1245 validation accuracy: 0.9726

Perentile: 96, threshold: 0.1383, validation accuracy: 0.9772

Perentile: 97, threshold: 0.1555, validation accuracy: 0.9795

Perentile: 98, threshold: 0.192, validation accuracy: 0.9795

Perentile: 99, threshold: 0.2614, validation accuracy: 0.8174

Best validation accuracy: 0.9795 for threshold: 0.1555.

---