Malware Scanner

Project Title

Malware Scanner

Project Area of Specialization Cyber SecurityProject Summary

Routine file usage is an essential aspect in daily tasks. Such tasks require malware free environment. An efficient machine learning-based Malware Scanner is proposed to scan any kind of virus or malicious software present in scanned file. This will prevent the user from the fatigue of scanning the file in different anti-virus programs and from buying subscriptions or paying for different programs for every purpose. The desktop application will detect the malicious file through machine learning. Machine learning techniques provide a promising solution in an effective manner by using data splitting. All that is needed is to install the application.

Project Objectives

The goal is to provide a state-of-the-art desktop application that is open-source and provides service better than the other paid applications in the relevant market. The proposed approach may provide a security of system by machine learning approach.
The concept of machine learning methods for malware detection is not new. Several types of studies were carried out in this field, aiming to figure the accuracy of different methods.
Discussed by (David aklan) in [4], the detection method based on modified Random Forest algorithm in combination with Information Gain for better feature representation.
Basically, a clean and malicious dataset of PE file may be used for training the dataset on the base of these two data. Algorithms may be varying such as Decision Tree, Naïve Bayes and KNN. The aim is to provide accuracy approximately 60-80% through this application and provide real-time protection that no one provided before.   

Project Implementation Method

We will be making a desktop application therefore, an agile method of adaptive software development (ASD) would be used as a software process model. Actual model of our project is as under

Benefits of the Project

Major techniques of detecting malware are Signature-based detection which traditional antivirus and is used to compare the known signature of viruses with target file [2]. This method is not accurate as if malware changes the file signature or antivirus do not have newly created virus signature (zero day), therefore antivirus misses such type of virus. It motivates towards using machine learning approach. This will provide user accurate result which will save their time and money. 

LIST OF AUDIENCE EFFECTED
Domestic users
Industries
Bank employees
Law firms
Government Services
Any organization or person who carry any important data

Technical Details of Final Deliverable

Main focus is to build the project from scratch with flexibility of changeability and reusability. Different tools and languages may be used. 
Languages:
Python
C#.Net
Tools:
Pycharm Ide
Anaconda (Jupyter Notebook)
Tkinter(Framework)
Thrift(server)

Final Deliverable of the Project Software SystemCore Industry ITOther IndustriesCore Technology NeuroTechOther TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	0
dataset	Miscellaneous	0	0	0