Aircraft Icing Severity Assessment Based on Machine Learning with Multi-Source Data Fusion

This dataset contains Pilot Reports (PIREPs) of aircraft icing events used in the study “Aircraft Icing Severity Assessment Based on Machine Learning with Multi-Source Data Fusion.” The dataset includes both icing and non-icing cases collected over multiple regions and time periods to ensure balanced sampling and robust model training. Sensitive information such as flight identifiers has been removed to preserve data privacy. The dataset can be used for research on aircraft icing detection, aviation weather forecasting, and machine learning-based atmospheric hazard assessment.