DMP: Air Pollution Emissions Classification using European Open Data

Dataset Description: Air Pollution Emissions Classification Dataset

Context and methodology

This dataset is used in a machine learning project analysing air pollution emissions in Europe. The data comes from the European Open Data Portal (EAA20 - Air Pollution Emissions dataset).

The purpose is to train a classification model that predicts pollution levels. Since the dataset has no labels, a binary target variable (high/low pollution) is created using a median threshold.

The dataset is downloaded in CSV format and processed using Python, including data cleaning and feature engineering

Technical details

• The dataset is organised as follows:

  • data/raw/ – original dataset
  • data/processed/– cleaned data
  • src/ – code
  • results/ – outputs
  • docs/ – documentation

  File naming follows consistent conventions (e.g., raw data, processed data, model files, and plots).

  The dataset can be used with Python (pandas, numpy, scikit-learn). Documentation and code are provided.

Further details

• The dataset is publicly available and reusable under its licence.
• No personal or sensitive data are included.
• The dataset is managed following FAIR principles to ensure that it is findable, accessible, interoperable, and reusable.