Accuracy Enhancement for Intrusion Detection Systems Using LSTM Approach

Post-pandemic threats to the network has shown that there is a need for lot of work to be done on the accuracy of IDS. Conventional research has only produced a limited number of viable options for efficient intrusion detection. When put into practice, the conclusions and suggestions that have been drawn from this research will have a considerable impact on the approach that is used to accurately predict intrusions. RNN-LSTM-based approach in the proposed model not only enhanced the accuracy of IDS, but yields less false positives and rapid detection of potential security threats. The results of the proposed IDS model have been compared with traditional IDS, where proposed model provides better Accuracy, Precision, Recall, and F1 Score. The suggested model trained on a substantial dataset, which significantly increased the possibility to provide comparatively better results. In order to improve IDS detection, it is recommended that future study continues to make use of the same paradigm of using deep learning methods.