Research Article

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

Table 4

The optimization results when the dimension is 50.

PSOBSAFOPSOHPSOASPSOOLPSOGEPSOHMaPSOAPPSO

F12.97E + 012.83E + 011.64E + 023.45E + 016.27E + 004.95E + 017.28E + 011.00E + 01
F26.92E-014.64E + 023.47E + 021.04E + 004.01E + 001.57E + 004.41E + 016.17E-01
F35.79E-043.41E + 026.26E + 011.12E + 006.73E + 011.78E-012.81E + 023.47E-04
F42.57E + 012.44E + 011.42E + 022.98E + 018.64E + 004.27E + 016.28E + 015.42E + 00
F53.10E-034.69E-012.97E-012.36E + 001.85E + 001.15E + 007.01E-011.99E-01
F61.56E-026.96E + 009.21E + 009.61E-033.53E + 021.84E + 012.22E + 003.18E-04
F76.10E + 011.68E + 038.43E + 026.74E + 001.64E + 035.93E + 013.29E + 026.81E + 01
F83.22E-012.19E + 005.61E + 006.76E + 001.60E + 012.49E-019.31E + 001.63E-02

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