🛡️ SafeClick — Phishing Link Detection & Learning System

Detect phishing links instantly. Learn why they're dangerous.

---

📖 About The Project

SafeClick is a machine learning-based phishing URL detection system that helps users identify whether a link is safe, suspicious, or a phishing attempt. It doesn't just detect — it educates users about phishing attacks and teaches them how to stay safe online.

The Problem

• 🎣 91% of cyber attacks start with a phishing email
• 📧 3.4 billion phishing emails are sent every day
• 💰 Average cost of a phishing attack: $4.76 million
• 😰 Most people can't tell a phishing link from a real one

Our Solution

SafeClick analyzes any URL using 36 intelligent features and a Random Forest ML model trained on 3,00,000+ real URLs to give users:

• A phishing score (0-100)
• A risk level (Safe / Suspicious / Phishing)
• Detailed reasons explaining why the URL is dangerous
• Educational tips about phishing and online safety

---

🎯 Features

Feature	Description
🔍 URL Scanner	Paste any URL to instantly check if it's phishing
📊 Phishing Score	Get a score from 0-100 (higher = more dangerous)
🚦 Risk Level	Clear labels: SAFE / SUSPICIOUS / PHISHING
📋 Detailed Reasons	Explains exactly WHY the URL is flagged
💡 Learn Section	Educational tips and facts about phishing
🤖 Machine Learning	Trained on 3,00,000+ real-world URLs
📈 94-95% Accuracy	Reliable and realistic detection
📱 Responsive Design	Works on desktop, tablet, and mobile
⚡ Fast Analysis	Results in under 1 second
🔄 Batch Scanning	Scan up to 10 URLs at once via API

---

🏗️ Project Structure

SafeClick/
│
├── 📁 backend/ ← Python (Flask + Machine Learning)
│ ├── 📁 api/
│ │ ├── init.py
│ │ ├── app.py ← Flask app with CORS
│ │ └── routes.py ← API endpoints (/api/scan)
│ │
│ ├── 📁 ml/
│ │ ├── init.py
│ │ ├── feature_extractor.py ← Extracts 36 features from URL
│ │ ├── predict.py ← Loads model & predicts score
│ │ ├── explainer.py ← Generates human-readable reasons
│ │ └── train_model.py ← Training pipeline
│ │
│ ├── 📁 models/
│ │ ├── phishing_model.pkl ← Trained ML model (generated)
│ │ ├── scaler.pkl ← Feature scaler (generated)
│ │ └── model_info.json ← Model metadata & metrics
│ │
│ ├── 📁 data/
│ │ ├── 📁 raw/ ← Raw dataset from Kaggle
│ │ ├── 📁 processed/ ← Cleaned & filtered dataset
│ │ ├── download_data.py ← Sample data generator
│ │ └── filter_dataset.py ← Filters 450K → 70K URLs
│ │
│ ├── 📁 notebooks/
│ │ ├── 01_data_exploration.ipynb
│ │ ├── 02_feature_engineering.ipynb
│ │ ├── 03_model_training.ipynb
│ │ └── 04_evaluation.ipynb
│ │
│ ├── 📁 tests/
│ │ ├── test_features.py ← Feature extraction tests
│ │ └── test_api.py ← API endpoint tests
│ │
│ ├── config.py ← All settings & constants
│ ├── requirements.txt ← Python dependencies
│ └── run.py ← Start backend server
│
├── 📁 frontend/ ← HTML + CSS + JavaScript
│ ├── index.html ← Main scanner page
│ ├── about.html ← About the project
│ ├── how-it-works.html ← How it works page
│ │
│ ├── 📁 css/
│ │ ├── style.css ← Global styles
│ │ ├── scanner.css ← Scanner section styles
│ │ └── results.css ← Results display styles
│ │
│ ├── 📁 js/
│ │ ├── api.js ← ⭐ Connects frontend to backend
│ │ ├── scanner.js ← Scanner form logic
│ │ ├── results.js ← Results display + animation
│ │ └── utils.js ← Helper functions + tips
│ │
│ └── 📁 assets/
│ └── 📁 images/
│
├── .gitignore
└── README.md

💻 Technology Stack

Layer	Technology	Purpose
Frontend	HTML5, CSS3, JavaScript	User interface & scanner
Backend	Python 3.8+, Flask	REST API server
ML Model	Scikit-learn, Random Forest	Phishing prediction
Data Processing	Pandas, NumPy	Dataset handling
Feature Extraction	urllib, regex, tldextract	URL analysis
Visualization	Matplotlib, Seaborn	Model evaluation charts
Communication	REST API (JSON)	Frontend ↔ Backend

---

🚀 Getting Started

Prerequisites

• Python 3.8+ — Download Python
• Web Browser — Chrome, Edge, or Firefox
• Git — Download Git

Installation

1. Clone the Repository

git clone https://github.com/pervej_rony27/SafeClick.git
cd SafeClick

Step 2: Setup Backend

cd backend

# Create virtual environment
python -m venv venv

# Activate virtual environment
# Windows:
venv\Scripts\activate
# Mac/Linux:
source venv/bin/activate

# Install all dependencies
pip install -r requirements.txt