Table 2 Other models’ performance
From: Modelling email traffic workloads with RNN and LSTM models
Email traffic | Technical University of Crete | Murdoch University | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
KNN | Random Forest | Linear regression | ARIMA | SARIMA | KNN | Random Forest | Linear regression | ARIMA | SARIMA | |
Incoming | ||||||||||
Users | 240.28 | 673.92 | 240.28 | 1697.90 | 99.99 | 71.46 | 74.06 | 82.33 | 469.21 | 198.00 |
System | 19.39 | 183.80 | 29.25 | 122.37 | 99.99 | 92.97 | 3181.38 | 92.98 | 27.13 | 133.66 |
Outgoing | ||||||||||
Users | 424.68 | 13436.85 | 424.68 | 14119.89 | 99.72 | 66.88 | 1254.35 | 66.85 | 409.85 | 840.70 |
System | 18.33 | 318.15 | 23.08 | 128.39 | 99.99 | 52.80 | 112.80 | 78.91 | 159.99 | 57.28 |
Spam | 28.84 | 554.92 | 28.80 | 567.08 | 99.89 | |||||