Phishing Detection System

A machine learning-powered phishing detection system built with FastAPI that detects suspicious text messages and malicious URLs.

The system analyzes user input and classifies it as either:

• Phishing
• Safe

This project combines Natural Language Processing (NLP), URL feature extraction, and machine learning models to improve phishing detection accuracy.

---

Features

• Detect phishing text messages
• Detect malicious URLs
• FastAPI REST API
• Machine learning integration
• TF-IDF text classification
• URL feature extraction
• Frontend integration support
• CORS enabled
• Swagger API documentation

---

Tech Stack

Backend

• Python
• FastAPI
• Uvicorn

Machine Learning

• Scikit-learn
• Logistic Regression
• TF-IDF Vectorizer
• DictVectorizer
• NLTK

Data Processing

• Pandas
• Regex
• Joblib

---

Project Structure

app/
│
├── main.py
│
├── routes/
│   └── predictor.py
│
├── services/
│   ├── prediction_service.py
│   ├── predict_text.py
│   └── predict_url.py
│
├── utils/
│   ├── helpers.py
│   └── url_features.py
│
├── schemas/
│   └── predict_schema.py
│
├── models/
│   ├── text_model.pkl
│   ├── text_vectorizer.pkl
│   ├── url_model.pkl
│   └── url_vectorizer.pkl
│
├── dataset/
│   ├── spam.csv
│   └── Dataset.csv
│
├── train_model.py
├── train_url_model.py
│
└── requirements.txt