Phishing Website Detection Using Machine Learning Overview Developed an end-to-end machine learni...Phishing Website Detection Using Machine Learning Overview Developed an end-to-end machine learni...
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
Nandani's avatar
Nandani Jha
• Aug 5
Phishing Website Detection Using Machine Learning
Overview
Developed an end-to-end machine learning solution to identify phishing websites and help users distinguish malicious URLs from legitimate ones. The project focused on improving cybersecurity by accurately classifying websites based on engineered URL and website features.
Problem
Phishing attacks are one of the most common forms of cybercrime, causing financial loss and data breaches. Manual detection is time-consuming and unreliable, making automated classification essential.
My Role
Collected and prepared the phishing dataset.
Cleaned and preprocessed the data using Python (Pandas & NumPy).
Performed Exploratory Data Analysis (EDA) to understand feature relationships.
Trained and compared multiple machine learning models, including Logistic Regression, Decision Tree, Random Forest, SVM, and XGBoost.
Evaluated model performance using Accuracy and ROC-AUC metrics.
Applied SHAP to explain feature importance and improve model interpretability.
Built a simple Flask web application for real-time phishing URL prediction.
Documented the workflow and published the project on GitHub.
Tools & Technologies
Python
Pandas
NumPy
Scikit-learn
XGBoost
SHAP
Flask
Jupyter Notebook
Git & GitHub
Results
Achieved 96.63% accuracy using the XGBoost model.
Reached a ROC-AUC score of 0.9936, demonstrating excellent classification performance.
Created an interpretable model using SHAP feature importance.
Deployed the trained model in a user-friendly Flask application for instant predictions.
Deliverables
Cleaned and processed dataset
Machine learning training pipeline
Performance comparison of multiple models
SHAP explainability visualizations
Flask web application
Complete source code and documentation
This project demonstrates my ability to clean complex datasets, perform data analysis, build high-performing machine learning models, explain model decisions, and deliver production-ready solutions.
Data AnalysisData ScienceWeb ScrapingGitJupyterXGBoost
Post image
Back to feed
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started