✈️ SkyInsight — Airlines Data Engineering Project

📌 Overview

SkyInsight is an end-to-end Data Engineering platform designed to help airlines improve:

• Passenger satisfaction
• Customer loyalty
• Operational efficiency
• Environmental sustainability

The project integrates BI, Machine Learning, NLP, and Computer Vision into a unified pipeline — from raw data ingestion to intelligent insights.

---

🏗️ Project Architecture

Raw Data → Data Cleaning → Data Warehouse → BI → ML / NLP / CV

• Data Sources: CSV datasets (loyalty, flights, satisfaction)
• ETL: Python / SQL transformations
• Storage: Structured datasets (cleaned & modeled)
• Analytics: Power BI dashboards
• AI Layer:
  • Machine Learning
  • Natural Language Processing (NLP)
  • Computer Vision

---

📂 Repository Structure

Airlines_Project/
│
├── 01_Presentations/      # Project presentations (slides, demos)
├── 02_Documentation/      # Project documentation & architecture
├── 03_Data_Raw/           # Raw datasets (CSV files)
├── 04_Data_Cleaned/       # Processed & cleaned datasets
├── 05_Scripts/            # ETL scripts (Python / SQL)
├── 06_Dashboards/         # Power BI dashboards & exports
├── 07_ML/                 # Machine Learning models & notebooks
├── 08_CNN/                # Computer Vision (CNN / YOLO models)
├── 09_Reports/            # Final reports & analysis
├── 10_NLP/                # (Upcoming) NLP pipelines & text analysis
│
├── .gitignore
├── README.md
└── yolo11n.pt             # Pretrained YOLO model

---

📊 Business Intelligence Objectives

• Loyalty Analytics

  • Customer Lifetime Value (CLV)
  • Loyalty segmentation & churn tracking

• Flight Performance

  • Distance, revenue, and utilization metrics

• Passenger Satisfaction

  • Analysis across multiple service features

• Sustainability Reporting

  • CO₂ emissions tracking
  • Fuel efficiency per route

---

🤖 Machine Learning Use Cases

• Satisfaction Classification

  • Predict satisfied vs dissatisfied passengers

• Churn Prediction

  • Identify customers likely to leave loyalty programs

• Customer Segmentation

  • Clustering (K-Means / DBSCAN)

• Route Optimization

  • Improve load factor and efficiency

• Carbon Emission Prediction

  • Estimate CO₂ per passenger

---

🧠 NLP (Upcoming - 10_NLP/)

• Sentiment Analysis (reviews & feedback)
• Topic Modeling (LDA / BERTopic)
• Aspect-Based Sentiment Analysis
• Keyword Extraction
• Automated report generation

---

👁️ Computer Vision (CNN)

Located in 08_CNN/:

• Cabin cleanliness classification
• Passenger crowd detection (YOLO)
• Aircraft anomaly detection
• Baggage handling quality control

---

🌍 Sustainability Goals Alignment

This project supports:

• SDG 9 → Industry & Innovation
• SDG 12 → Responsible Consumption
• SDG 13 → Climate Action

---

⚙️ Tech Stack

• Languages: Python, SQL
• Data Processing: Pandas, PySpark
• Visualization: Power BI
• ML/DL: Scikit-learn, XGBoost, TensorFlow / PyTorch
• NLP: NLTK, SpaCy, BERTopic
• Computer Vision: OpenCV, YOLO
• Version Control: Git & GitHub

---

🚀 Getting Started

Clone the repository

git clone https://github.com/Larousse2001/Airlines_Project.git
cd Airlines_Project

📅 Project Roadmap

• ✅ Data Engineering Pipeline
• ✅ Data Warehouse & BI
• ✅ Machine Learning Models
• 🔄 NLP Module (10_NLP)
• 🔄 Advanced Computer Vision
• 🔄 Deployment (API / Dashboard)

---

👨‍💻 Contributors

• Achref AROUS
• Oumaima ROUIS
• Iheb KOUKI
• Mehdi JOUDI
• Eya CHIIBNI

---

📜 License

This project is for academic and educational purposes.