ritusingh0110/AI-Phishing-Detection-System

# AI-Powered Phishing Detection & Threat Intelligence Platform ## Project Overview This project is an AI-based cybersecurity system designed to detect phishing emails using Machine Learning, NLP, and cybersecurity threat analysis techniques. The platform analyzes email content, extracts suspicious indicators, calculates threat scores, and explains why an email is considered malicious. # Problem Statement Phishing attacks are one of the most common cybersecurity threats used to steal credentials, financial information, and sensitive user data. Traditional spam filters often fail to detect sophisticated phishing attempts. This project aims to improve phishing detection using: - Machine Learning - NLP - Cybersecurity feature engineering - Risk scoring - Explainable AI # Features - Phishing Email Detection - NLP-based Text Classification - TF-IDF Feature Extraction - URL Detection - IP Address Detection - Suspicious Keyword Analysis - Risk Score Calculation - Explainable AI Reasoning - Interactive Streamlit Dashboard - Docker Deployment Support # Tech Stack ## Machine Learning - Scikit-learn - RandomForestClassifier - TF-IDF Vectorization ## Frontend - Streamlit ## Backend - Python ## Libraries - Pandas - NumPy - Scipy - Joblib - Regex # System Architecture User Input ↓ Streamlit Frontend ↓ Feature Extraction Engine ↓ ML Classification Model ↓ Cybersecurity Risk Analysis ↓ Explainable AI Layer ↓ Prediction Output # Installation pip install -r requirements.txt # Run Application streamlit run app/app.py # Example Detection Input: URGENT! Verify your password immediately. Output: - Phishing Email - High Risk Score - Suspicious URL Detection # Future Improvements - VirusTotal API Integration - Deep Learning Models - Browser Extension - Real-Time Email Monitoring - SIEM Integration - Threat Intelligence APIs # Screenshots (Add Streamlit screenshots here) # Author Ritu Singh