Inspiration

Ransomware has become one of the most dangerous cybersecurity threats, causing data loss and financial damage worldwide. Traditional antivirus solutions often struggle to detect new ransomware variants. This inspired us to develop an AI-powered system that can intelligently analyze activities and identify potential threats.

Problems

• Increasing ransomware attacks.
• Traditional signature-based detection limitations.
• Difficulty detecting unknown attacks.
• Lack of simple tools for real-time threat monitoring.

What It Does

Our AI-Based Ransomware Detection System uses a Random Forest Machine Learning model to classify activities into:

• SS – Safe
• S – Suspicious
• A – Attack

The system provides a web dashboard for threat prediction, activity monitoring, security logs, and graphical threat visualization.

How We Built It

• Collected and preprocessed ransomware dataset.
• Applied Label Encoding and data cleaning.
• Split the data into 80% training and 20% testing.
• Trained a Random Forest Classifier using Scikit-Learn.
• Achieved 99.39% accuracy.
• Integrated the model with a Flask-based dashboard using HTML, CSS, JavaScript, and Chart.js.

Challenges We Faced

• Preparing and understanding the dataset.
• Encoding categorical features.
• Integrating the AI model with the web application.
• Designing an interactive dashboard within the project timeline.

What We Learned

Through this project, we gained practical experience in Machine Learning, cybersecurity, data preprocessing, Flask development, and integrating AI models into real-world applications.

Future Improvements

• Real-time file and network monitoring.
• User authentication and database support.
• Email alerts and cloud deployment.
• Advanced AI and Deep Learning models.

Built With

• chart.js
• css
• flask
• html
• javascript
• joblib
• numpy
• pandas
• python
• scikit-learn