Machine Learning-Based Trojan Detection for Enhanced Cybersecurity

This project focuses on detecting Trojan malware in network traffic using machine learning techniques. Trojans are malicious programs disguised as legitimate software, posing serious threats such as data theft, financial loss, and system compromise. To address this, we developed an intelligent detection system that analyzes network data and identifies suspicious patterns.

The dataset consists of 199 records with 86 features, including network flow characteristics such as packet length, protocol type, and flow duration. Data preprocessing steps were applied to clean, standardize, and encode the dataset. Feature selection and outlier detection techniques were used to improve model performance and ensure data quality.

A machine learning model, specifically a decision tree classifier, was implemented and trained on the processed data. The model achieved high accuracy (around 95%) in detecting Trojan activity. Performance was evaluated using metrics such as accuracy, precision, recall, and F1-score.

The results demonstrate that machine learning can effectively automate Trojan detection, enabling faster response times and reducing potential cybersecurity risks. This approach enhances system security, minimizes manual effort, and provides a scalable solution for evolving cyber threats.