In this session, you will work on one of two tasks focused on detecting threats using machine learning. Your goal is to analyze data and build a predictive model.
You can select one of the following:
Malware (malicious software) is any program or file designed to harm, exploit, or otherwise compromise a computer system, network, or user data. It includes viruses, worms, Trojans, ransomware, and spyware, often spreading through malicious downloads, email attachments, or software vulnerabilities.
Attackers use malware to steal sensitive information, disrupt operations, or gain unauthorized access to systems. Detecting malware is crucial for cybersecurity, as it helps prevent data breaches, financial losses, and system damage.
If you feel extra, you can create a web interface where users can upload a .exe file to check whether it is malicious or safe.
Develop a model to determine whether a .exe file is malicious or safe based on its characteristics.
- Extract features from executable files.
- Train a machine learning model to classify them.
- Evaluate your model’s accuracy and performance.
A trained model that, when given a new .exe file, predicts if it is malicious or legitimate.
Phishing is a form of cyber fraud where attackers impersonate trusted companies or individuals to steal sensitive information, such as login credentials or account details. This is often done through emails or other communication channels.
Attackers widely use phishing because it is easier to trick someone into clicking a malicious link that looks legitimate than to bypass a computer’s security defenses.
The main goal is to build a system that identifies whether a given URL is a legit or fake website created to perform fraud. If you feel extra, you can create a web interface to allow users to input a URL and receive a real-time prediction on whether it is phishing or legitimate.
This interface can enhance usability by providing an intuitive way for users to test URLs without needing to run code manually. You can implement it using frameworks like Flask or Streamlit for a simple and interactive experience.
- Analyze URL structures and extract key attributes.
- Train a classification model to detect phishing attempts.
- Test and evaluate your model’s effectiveness.
A model that classifies URLs as phishing or safe based on extracted features.
- Clone this repository.
- Choose one of the two tasks.
- Follow the provided dataset and guidelines.
- Build and evaluate your model.
For those interested in a more advanced challenge, consider integrating both tasks—malware detection and phishing URL prediction—into a single platform for comprehensive cybersecurity analysis.
💡 This session is hands-on—feel free to collaborate, ask questions, and experiment!
👨💻 Happy coding & good luck!