This project is based on two-layer neural network system to identify normal or anomalous activities by training and validating the data in our model.
The accuracy: 1st epoch: around 40% last epoch: around 90% Final accuracy: in the range of 85-95%.
Data from: Bhosale, Sampada. “Network Intrusion Detection.” RSNA Pneumonia Detection Challenge | Kaggle, 9 Oct. 2018, www.kaggle.com/sampadab17/network-intrusion-detection#Test_data.csv. 25,192 rows * 41 columns