Unsw Nb-15dataset is a benchmark dataset used for network intrusion detection research. It was created by the Australian Centre for Cyber Security (ACCS) at the University of New South Wales (UNSW) in 2015. The dataset is designed to simulate real-world network traffic and to help researchers develop, test, and evaluate intrusion detection systems (IDS) and anomaly detection models.
The dataset is widely used in the academic community for cybersecurity-related research, particularly in building models that can identify and detect network intrusions or abnormal activities in network traffic.
Assignment:- Find Attack class , accuracy score, precision score, F1 score, ROC and FPR using Decision Tree, Random Forest and Artificial Neural Network. Here is an python file which comprises of ML and DL algorithms which are used on the training and testing sets of UNSW-NB15 dataset.
ML algorithms used:- Decision Tree Classifier, Random Forest Classifier
DL algorithms used:- Artificial Neural Network
"attack_cat" aka Attack Class of the Dataset of UNSW NB-15 contains different elements which are normal and anomalies. The attack class are counted using dataset.values_count(); funstion of pandas library.The Elements of Attack class are as follows:-
Normal 37000
Generic 18871
Exploits 11132
Fuzzers 6062
DoS 4089
Reconnaissance 3496
Analysis 677
Backdoor 583
Shellcode 378
Worms 44
