Scribd
Upload
27 views8 pages

ML Malware Detection Hackathon

The document outlines a hackathon challenge where participants must develop a machine learning model to classify malware into specific categories using a provided dataset. Key challenges include designing a robust and scalable model that generalizes to unseen samples, with a focus on various malware types. Deliverables include the model, implementation code, a detailed report, and a presentation of findings.

Uploaded by

SAKETA SRI RAMACHARYULU GUDIMELLA (RA2211026010111)
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
27 views8 pages

ML Malware Detection Hackathon

The document outlines a hackathon challenge where participants must develop a machine learning model to classify malware into specific categories using a provided dataset. Key challenges include designing a robust and scalable model that generalizes to unseen samples, with a focus on various malware types. Deliverables include the model, implementation code, a detailed report, and a presentation of findings.

Uploaded by

SAKETA SRI RAMACHARYULU GUDIMELLA (RA2211026010111)
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Securing Systems using ML/AI -

Malware Detection

Mined Hackathon 2025 -


Nirma University
Agenda

● Objective
● Challenges
● Dataset explanation
● Deliverables

Crest Data Systems Confidential. Do NOT Distribute.


Objective

● Participants are tasked with building a machine learning model to classify malware into predefined
categories using the provided dataset.

● The goal is to develop efficient system that can accurately identify the type of malware based on the
given data.

● The solution will help enhance the detection and prevention mechanisms for real-world cybersecurity
systems.

Crest Data Systems Confidential. Do NOT Distribute.


Challenges

Participants must design, train, and evaluate a machine learning model to classify malware into one of the
following categories:

● Benign = 0
● RedLineStealer = 1
● Downloader = 2
● RAT = 3
● BankingTrojan = 4
● SnakeKeyLogger = 5
● Spyware = 6

The implementation must be robust and scalable, with a focus on generalizing to unseen malware samples.

Crest Data Systems Confidential. Do NOT Distribute.


Dataset Explanation

The overall features are distributed in three sections:

- Portable executable: It contains 52 fields of PE headers, 9 field values of 10 PE section,


- DLL imported: contains the DLLs imported by each malware family.
- API functions: contains the API functions called by these malware

Dataset: https://drive.google.com/drive/folders/17BKEb8ujyf1lpX2hHCcXrl2Zc7mb2Vmp?usp=drive_link

Crest Data Systems Confidential. Do NOT Distribute.


Dataset Explanation

To understand the different part of dataset:

Portable executable:

> https://stixproject.github.io/data-model/1.2/WinExecutableFileObj/DOSHeaderType/
> https://learn.microsoft.com/en-us/windows/win32/debug/pe-format
> https://0xrick.github.io/win-internals/pe5/#sections-and-section-headers
DLL imported:

> It represents 629 type of DLL files are used for the respective executable.

API functions:

> It represents 21918 type of API function call done from the respective executable.

Crest Data Systems Confidential. Do NOT Distribute.


Deliverables

● The machine learning model and its implementation code.


● A detailed report explaining:
○ Model architecture and approach.
○ Preprocessing and feature engineering techniques used.
○ Challenges faced and how they were overcome.
● A presentation of your solution, including key findings and potential areas for improvement.
● Test.csv along with predictions
● Trained model file

Crest Data Systems Confidential. Do NOT Distribute.


THANK YOU!

[email protected]
https://www.crestdata.ai

Crest Data Confidential. Do Not Distribute.

You might also like