Anomaly Detection For Network Security

Network security is critical these days as network technology advances quickly and internet technology advances even faster, particularly since network threats increase daily. Network anomaly detection is one of the techniques that can be used to protect the security of a network. There has been a lot of recent research on approaches for detecting anomalies. This article presents a unique method for detecting network anomalies that makes use of the Autoencoder model. This proposed method has been shown to be applicable to time series data. The traffic data is determined to be normal or anomaly by comparing the reconstruction error to the threshold value. Referring to the real modern large scale network traffic data, UNSW-NB 15 and CICIDS2017 data set is chosen to evaluate the performance of the proposed Autoencoder model