Malware Prediction System using Machine and Deep Learning

This project is based on automated malware prediction. The main aim of this project is to keep defenders a step ahead of attackers, an evolution algorithm has been implemented that predicts pattern of future malware behavior. We?ll focus on multiple aspects of the automated malware prediction by mac

Project Title

Malware Prediction System using Machine and Deep Learning

Project Area of Specialization

Artificial Intelligence

Project Summary

This project is based on automated malware prediction. The main aim of this project is to keep defenders a step ahead of attackers, an evolution algorithm has been implemented that predicts pattern of future malware behavior. We’ll focus on multiple aspects of the automated malware prediction by machine learning and deep learning technique in order to achieve maximum accuracy. In the first part of our project, we have to analyse the best features of the dataset, and used different machine learning based methods that provides the best prediction accuracy. The models are designed on the top of Google Colab and Jupyter Notebook. In second part of the project we have to Implement Artificial Neural Networks to predict the attacks. We experimented with adding different numbers of system features and hidden neural layers to predict the accuracy of the model. For training the models we’ll use Light GBM gradient boosting frameworks.

Project Objectives

• Traditional AV or anti malware solution works on signatures.
• Polymorphic malware bypass these traditional detection solutions.
• Malware detection is a classification problem.
• Machine learning learns the dataset and then based on its learning, it predicts: malware/not malware.
• Supply more and more latest dataset to improve machine learning model performance.

Project Implementation Method

1. Implementation of all dependencies.
2. Adding and training the dataset models
3. Data Preprocessing.
4. Implementing of Machine learning models.
5. Implementing of Neural Networks models.
6. Implementation of Gradient Boosting Framework.
7. Comparison Of different methods based on accuracy.
8. Implementing all the model on flask framework.
9. Detecting the malware files as well as predicting it's types and what is the solution of that malware.

Benefits of the Project

• Traditional AV or anti malware solution works on signatures.
• Polymorphic malware bypass these traditional detection solutions.
• Malware detection is a classification problem.
• Machine learning learns the dataset and then based on its learning, it predicts: malware/not malware.
• Supply more and more latest dataset to improve machine learning model performance.

Technical Details of Final Deliverable

Programming Language : Python
Web Framework : Flask Scripts
Algorithms Used : Random Forest, Decision Tree & GBA
Deep Learning Model : Neural Network
Front End : HTML, CSS
IDE : Jupyter Lab, PyCharm, Visual Studio Code

Final Deliverable of the Project

Software System

Core Industry

IT

Other Industries

Education

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Responsible Consumption and Production

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com