Mariben-Summarizer

A fullstack application that automatically analyzes articles using two AI services.

• Hugging Face : Zero-Shot Classification

• Google Gemini : Summarization, Sentiment Analysis, and Contextual Synthesis

---

Project Structure

.
├── app/                    
│   ├── __init__.py
│   ├── database.py
│   ├── dockerignore
│   ├── Dockerfile
│   ├── hasher.py
│   ├── main.py             
│   ├── models.py
│   └── schema.py
│
├── hybrid-app/              
│   ├── app/
│   ├── public/
│   ├── Dockerfile
│   ├── package.json
│   ├── next.config.ts
│   ├── tailwind / postcss
│   └── README.md
│
├── services/               
│
├── docker-compose.yml      
├── .env                    
├── requirements.txt         
├── pytest.ini
└── README.md               

---

Features

• Automatic article analysis

• Zero-shot text classification

• AI-generated summaries

• Tone and sentiment detection

• Secure JWT authentication

• Frontend ↔ Backend API communication

• Docker & Docker Compose support

---

Tech Stack

Backend

• Python

• FastAPI

• JWT Authentication

• PostgreSQL

• Test:Mock

Frontend

• Next.js

• Tailwind CSS

DevOps

• Docker

• Docker Compose

---

Installation & Setup

1- Clone the repository

git clone https://github.com/mariambenali/Hybrid-Analyzer
cd Hybrid-Analyzer

2- Run with Docker

docker-compose up --build

• Frontend: http://localhost:3000

• Backend API: http://localhost:8000

3- Run without Docker

Backend:

python -m venv myvenv
source myvenv/bin/activate
pip install -r requirements.txt
uvicorn app.main:app --reload

Frontend:

cd hybrid-app
npm install
npm run dev

---

Testing

pytest

---

👩‍💻 Author

Miriam Benali GitHub: https://github.com/mariambenali