Detection of IT and Mobile Network Attacks

The project focuses on the detection of IT and mobile network attacks using machine learning techniques. With the increasing complexity of network systems, identifying malicious activity has become crucial. This study uses the Edge-IIoTset dataset to train models capable of detecting multiple attack types, including DDoS_UDP, DDoS_ICMP, DDoS_TCP, Port_Scanning, Ransomware, and Begin. The system is implemented using Python and the Flask framework, providing modules for registration, login, dataset upload, model evaluation, prediction, and viewing results. Machine learning algorithms such as Support Vector Machine (SVM), Random Forest, and XGBoost are applied to classify network activity accurately. The project aims to reduce manual effort in monitoring network traffic and improve detection accuracy. Evaluation metrics such as accuracy, precision, recall, and F1-score are used to measure model performance. Users can visualize predictions and model results through a simple interface. Overall, the project demonstrates an effective approach for automated detection of network attacks and provides a platform for testing and evaluation of different machine learning models.

Keywords: network attacks, IT security, mobile network, machine learning, dataset analysis, Flask, Python, prediction, model evaluation, attack detection