Malicious Content Detection using CNN

Malware is a malicious code that was developed with an aim of harming and disrupting a computer or computer network. Malware malicious content is distributed through the internet resulting in corruption and loss of data of hundreds of millions of people around the globe. Although, many methods have

Project Title

Malicious Content Detection using CNN

Project Area of Specialization

Cyber Security

Project Summary

Malware is a malicious code that was developed with an aim of harming and disrupting a computer or computer network. Malware malicious content is distributed through the internet resulting in corruption and loss of data of hundreds of millions of people around the globe. Although, many methods have been proposed to detect malicious content detection and it turns out to be increasingly more difficult for those techniques to give a delightful outcome nowadays. The machine learning method is becoming popular in the detection of malicious content, however, most of the existing machine-learning algorithm uses the shallow learning algorithm like (SVM). Recently, a deep learning CNN algorithm has shown a superior result which inspires us to implement our proposed idea. The proposed method uses a Convolutional Neural Network which is a class of a deep neural network as a classification algorithm. We are basically working on research work to build software that will able to detect malicious content over the traffic using the deep learning method. The basic idea is to train the model on big data using NVIDIA GPUS. We are aiming for the highest accuracy detection, and we are hopeful that, we will able to achieve our objective. 

Project Objectives

The world’s dependency on the internet is ever-growing. Information is the most valuable resource in the world. To protect information from unwanted hands is of utmost importance. Information gets stolen when a network gets breached. Therefore, our research work proposed a solution to this problem. The project objective is to detect malicious content over the network traffic with high accuracy.

Project Implementation Method

We will be building the predictive model in the following steps:

1: Since the Convolutional Neural Network model feed data as an image, therefore, we first Convert each malicious content (malware) to a grayscale image to feed into the model. We are using the CICIDS2017 dataset. The most updated dataset.

2: Extract and Engineer Features from grayscale images

3: Build CNN network architecture

4: Compile, fit and train the model.

Benefits of the Project

• Benifits of  the PROJECT
• Monitor and evaluate threats, catch intruders and take action in real-time to thwart such instances that firewall or antivirus software may miss.
• Prevent DoS/DDoS attacks.
• Maintain the privacy of users as IPS records the network activity only when it finds an activity that matches the list of known malicious activities.
• Stop attacks on the SSL protocol or prevent attempts to find open ports on specific hosts.
• Detect and foil OS fingerprinting attempts that hackers use to find out the OS of the target system to launch specific exploits.

Technical Details of Final Deliverable

Our project started on 5, October 2020.  In the next three months, we will do a literature review till December 2020. During the month of January,2021,  preprocessing and feature extraction will be done. From the month of February 2021 to April 2021, training data will start on NVIDIA GPUS. In the month of may experimentation and testing. Finally, the last 25 days for the research project report. 

Final Deliverable of the Project

Software System

Core Industry

Security

Other Industries

IT

Core Technology

Artificial Intelligence(AI)

Other Technologies

Big Data

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

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