NETWORK TRAFFIC INTRUSION DETECTION ATTACK TYPE CLASSIFICATION

Abstract

With the rapid expansion of internet usage and network-based applications, cyber
threats and network attacks have increased significantly, posing serious risks to data
security and system integrity. Traditional security mechanisms such as firewalls and
signature-based intrusion detection systems are often limited in their ability to detect
new, unknown, or evolving attack patterns. Therefore, there is a growing need for
intelligent and adaptive systems that can effectively monitor network traffic and
identify malicious activities in real time.
This project, titled “Network Traffic Intrusion Detection Attack Type Classification,”
proposes a machine learning-based approach to detect and classify different types of
network intrusions. The system analyzes network traffic data using publicly available
datasets such as NSL-KDD or CICIDS2017, which contain detailed information about
network connections and various attack scenarios. Key features such as packet
statistics, protocol types, and connection attributes are extracted and used as input for
training the models.
Data preprocessing techniques, including normalization, feature selection, and
encoding, are applied to enhance data quality and model performance. Machine
learning algorithms such as Random Forest, Support Vector Machine (SVM), and KNearest
Neighbors (KNN) are implemented to classify network traffic into normal or
attack categories. Furthermore, the system identifies specific types of attacks such as
Denial of Service (DoS), Probe, Remote to Local (R2L), and User to Root (U2R).