Shreya934-bot/Malicious-URL-Threat-Detector

# 🛡️ Malicious URL Threat Detector

    

## 🚀 Live Demo 🔗 **Hugging Face App:** https://shreya934-hybrid-url-detector.hf.space 🔗 **Hugging Face Space:** https://huggingface.co/spaces/shreya934/hybrid-url-detector # 📌 Project Overview Malicious URL Threat Detector is an AI-powered cybersecurity application that detects malicious URLs using a combination of: - 🧠 Deep Learning (BiLSTM) - ⚡ Hybrid Rule-Based Threat Detection - 🔍 Feature Engineering - 🌐 URL Intelligence Analysis The system classifies URLs into: - ✅ Benign - ⚠️ Phishing - 🚨 Malware - 🔥 Defacement This project combines machine learning intelligence with handcrafted cybersecurity logic to improve detection performance and real-world reliability. # ✨ Features ## 🔍 Threat Detection - Phishing URL Detection - Malware URL Detection - Website Defacement Detection - Benign URL Classification ## 🧠 AI Features - BiLSTM Deep Learning Architecture - NLP Tokenization - URL Sequence Processing - Threat Probability Analysis ## 📊 Visualization & Analytics - Interactive Threat Distribution Charts - URL Intelligence Dashboard - Feature Analysis Panel - Threat Risk Scoring ## ⚡ Additional Features - Downloadable Threat Reports - Scan History Tracking - Quick Test URL Simulation - Responsive Cybersecurity UI - Live Threat Analysis # 🧠 Model Architecture The deep learning model uses: - Embedding Layer - Bidirectional LSTM - Dense Neural Layers - Dropout Regularization - Softmax Multi-Class Classification ### Hybrid Detection Logic The application also integrates rule-based threat analysis for: - suspicious keywords - executable file detection - PHP defacement patterns - suspicious TLDs - IP-based URLs # 🛠️ Tech Stack | Technology | Purpose | |---|---| | Python | Core Development | | TensorFlow / Keras | Deep Learning | | Gradio | Web Application | | Plotly | Data Visualization | | NumPy | Numerical Computing | | Pandas | Data Handling | | Scikit-learn | Label Encoding | | NLP Tokenization | URL Processing | # 📂 Project Structure Malicious-URL-Threat-Detector/ │ ├── app.py ├── requirements.txt ├── tokenizer.json ├── label_encoder.pkl ├── bilstm_weights.weights.h5 │ ├── screenshots/ │ ├── homepage.png │ ├── benign_detection.png │ ├── phishing_chart.png │ ├── malware_chart.png │ └── feature_analysis.png │ └── assets/ # 📸 Screenshots ## 🏠 Homepage  ## ✅ Benign URL Detection  ## 🔍 Feature Analysis  ## 📜 Scan History & Download Report  # ⚙️ Installation ## 1️⃣ Clone Repository git clone https://github.com/Shreya934-bot/Malicious-URL-Threat-Detector.git ## 2️⃣ Move Into Project Folder cd Malicious-URL-Threat-Detector ## 3️⃣ Install Requirements pip install -r requirements.txt ## 4️⃣ Run Application python app.py # 🌐 Deployment This project is deployed using: - Hugging Face Spaces - Gradio Framework # 📈 Model Performance ✅ Test Accuracy: **87%** ✅ Multi-Class URL Classification ✅ Hybrid AI + Rule-Based Detection ✅ Real-Time Threat Analysis # 🔮 Future Improvements - 🌐 Real-Time Threat Intelligence APIs - 🧩 Browser Extension Integration - 🤖 Transformer-Based Detection Models - ☁️ Cloud Database Logging - 📡 Live Domain Reputation Scanning - 🔐 Advanced Threat Intelligence Integration # 👩‍💻 Author ## Shreya Verma 🎓 B.Tech CSE (AI & ML) 🛡️ Cybersecurity & AI Enthusiast 💻 Deep Learning Developer # ⭐ If You Like This Project Give this repository a ⭐ on GitHub! # 📜 License This project is licensed under the MIT License.