Network Anomaly Detection Using a Hybrid Machine Learning Framework
Prerak Sudan · 2024
Network anomaly detection system (NADS) is widely used in a variety of sectors and allows for the monitoring of computer networks that react differently from the network protocol. Yet, the issue comes when several application areas possess various defining environmental abnormalities. These factors make it tough and complex to select the optimal algorithms that fit and satisfy the criteria of particular domains. Furthermore, the problem of centralization might result in the catastrophic collapse of a network when strong malicious software is introduced. We therefore offer a novel hybrid machine learning (ML) approach for NADS in this study termed long/short-term memory and support vector machine (LSTM-SVM). The results of the experiments demonstrate that, in terms of identifying network abnormalities, the suggested LSTM-SVM method outperforms all other methods currently in use.