Air Paradis Project – Sentiment Analysis

📘 This project is also available in French 🇫🇷

This repository contains the notebooks and source code for the sentiment analysis project on tweets, carried out as part of the OpenClassrooms "Air Paradis" program.

---

Business Context

The goal is to predict whether a tweet, or a short written expression, is perceived as positive or negative, in the context of an airline company.
The project is based on the public Sentiment140 dataset.

---

Project Description

The project aims to develop a complete sentiment analysis system, including:

• Exploratory data analysis of text data,
• Comparison of several classification models (classical, advanced, and state-of-the-art),
• Deployment of a REST API exposing an advanced model (based on embeddings and LSTM),
• Implementation of a CI/CD pipeline to automate testing and deployment,
• Local hosting on a NAS server to ensure data sovereignty.

---

Repository Structure

• notebooks/: analysis and modeling notebooks,
• deployment/: API code, tests, Dockerfile and deployment-related files,
• docs/: documentation and related materials,
• ../.github/workflows/: CI/CD pipelines.

---

Model Training

Training phases were mainly conducted on Google Colab to leverage GPU resources required for deep learning models.

---

Usage

The notebooks detail the analysis and training steps. The API is available in deployment/api.

---

Notes

• The requirements.txt files in notebooks/ come from Google Colab environments and may contain dependencies that are not optimal for local use.
• The .env file, excluded from the repository, contains the necessary environment variables.

---

This README will be updated progressively as the project advances.
Some code comments can remain in French (as well as notebook markdowns). All will be translated into English later.

---

Project completed as part of the OpenClassrooms "Data Scientist – Project 7" program