Aim

             To develop a real-time ransomware detection system using API call temporal intervals, enabling simulation and classification of ransomware behavior with a live interface.

Abstract

          This project presents a real-time ransomware detection system based on API temporal interval patterns. It simulates ransomware and benign activities using safe dummy API sequences and continuously monitors system behavior to detect threats using machine learning. A Tkinter-based GUI provides a user-friendly interface with login authentication, behavior simulation, and real-time classification. The detection model is trained on real API behavior logs, extracted from an existing dataset. The system is designed to mimic the behavior of active ransomware in a safe and controlled environment, demonstrating improved detection capabilities over existing approaches.

Proposed System

           The proposed system introduces a real-time ransomware detection framework. It includes:- Simulation of benign and ransomware behavior using realistic API patterns.- A GUI built with Tkinter, including user authentication. Continuous monitoring of API activity and classification using trained machine learning models.

              The system utilizes models such as: Light Gradient Boosting Machine (LightGBM), XGBoost, Random Forest, Gradient Boosting Classifier. These models are trained on ransomware and benign samples an best model is finalized. The system shows improved detection reliability and responsiveness.

Advantage

• Ransomware simulation and classification
• No actual malware used — only behavioral mimicry
• Safe and reproducible for demonstrations or analysis
• Accurate and fast detection with improved machine learning models
• User-friendly interface with login system