Modern Data Analytics - KU Leuven

Welcome to team Chad 👋

This is our project for the course Modern Data Analytics, where our objective is to develop an application that predicts the noise level in Naamsestraat, Leuven. The prediction model will be based on forecast weather and air quality data. By utilising machine learning models, we aim to provide valuable insights into the noise levels in the city, enabling residents and authorities to better understand and manage noise pollution.

$\mathbf{Members:}$

Jeh Mattummal $\mathbf{(r0861984)}$

Sven Nelles $\mathbf{(r0874870)}$

Jef Winant $\mathbf{(r0931958)}$

Yixin Mei $\mathbf{(r0911558)}$

Duc Tien Do $\mathbf{(r0916083)}$

Anh Phuong Dinh $\mathbf{(r0913033)}$

🌦 Data collection

• Air quality data:
  • Historical: scraped using OpenWeatherMap API
  • Forecast: scraped using Open-Meteo forecast air quality API
• Weather data:
  • Historical: scraped using Open-Meteo historical weather API with query
  • Forecast: scraped using Open-Meteo forecast weather API with query

📚 File organization

📦 
├─ README.md
├─ __pycache__
├─ conda_requirements.txt
├─ pip_requirements.txt 
├─ app
│  ├─ .streamlit
│  │  └─ config.toml
│  ├─ __pycache__
│  │  ├─ historical_noise.cpython-39.pyc
│  │  ├─ prediction_noise.cpython-39.pyc
│  │  └─ weather.cpython-39.pyc
│  ├─ historical_noise.py
│  ├─ main.py
│  ├─ prediction_noise.py
│  ├─ requirements.txt
│  ├─ weather.py
│  └─ woise-logo.png
├─ data
│  ├─ file40
│  ├─ file41.csv
│  ├─ file41
│  ├─ processed_air_quality_data.csv
│  ├─ processed_file40_data.csv
│  ├─ processed_file41_data.csv
│  ├─ processed_file42_data.csv
│  └─ processed_weather_data_leuven.csv
├─ model
│  ├─ model_noise_level_file40
│  ├─ model_noise_level_file42
│  └─ noise_types
└─ notebook
   ├─ 1_EDA.ipynb
   ├─ 2_scrape_and_process_data.ipynb
   ├─ 3_model_predict_noise_level_file40.ipynb
   ├─ 4_model_predict_noise_level_file42.ipynb
   ├─ 5_model_predict_noise_types.ipynb
   └─ 6_test_predictions.ipynb

⚙️ Installation guide

To set up the project environment, follow these instructions:

Clone the project repository

git clone https://github.com/aphdinh/MDA_KUL.git

Install the dependencies

python -m venv mda_chad
source mda_chad/bin/activate  # Windows: \venv\scripts\activate
pip install -r pip_requirements.txt

Navigate into the "app" folder using the cd command:

cd app

then run the app locally:

streamlit run main.py

The app has also been deployed here.