Research Article

Network Traffic Anomaly Detection Model Based on Feature Reduction and Bidirectional LSTM Neural Network Optimization

Table 3

The optimization results when the dimension is 5.

PSOBSAFOPSOHPSOASPSOOLPSOGEPSOHMaPSOAPPSO

F15.82E-148.16E-055.34E-054.03E-071.35E-062.31E-049.82E-112.88E-13
F22.16E+019.99E-021.41E-021.63E-021.24E+003.82E-032.20E-034.91E-12
F31.53E-123.39E-022.63E-041.88E-065.27E-015.79E-042.32E-022.94E-13
F41.63E-022.23E + 002.49E-021.96E + 002.07E + 003.99E-011.13E + 009.76E-06
F55.51E-132.32E-047.40E-052.46E-063.22E-021.07E-041.33E-131.04E-10
F60.09270.34920.11890.03007.44480.68660.01360.0043
F73.04E-013.92E + 001.28E + 002.37E + 006.67E + 007.16E-018.64E-026.38E-02
F83.88E-065.01E-027.30E-025.10E-031.07E-013.16E-022.38E-045.38E-07

Bold values signify the result is the optimal value for the corresponding function.