Visibility nowcasting in South Korea: a machine learning approach to class imbalance and distribution shift
arXiv:2605.21507v1 Announce Type: cross Abstract: Atmospheric visibility is a critical variable for transportation safety and air quality management, however, accurate prediction remains challenging due to the complex interactions between meteorological conditions and air pollutants, as well as the rarity of low-visibility events. This study introduces a machine learning framework to nowcast visibility in six major South Korean cities. To handle the imbalance in the 2018-2020 training data, we applied the Synthetic Minority Over-sampling Technique with Nominal and Continuous (SMOTENC) and Cond
The increasing availability of meteorological data and advancements in machine learning techniques, particularly for imbalanced datasets, enable more sophisticated atmospheric predictions.
Accurate visibility nowcasting is crucial for transportation safety and air quality management, directly impacting infrastructure and public health.
This research provides a more robust and localized method for predicting atmospheric visibility, potentially leading to improved operational decisions and safety protocols in specific regions.
- · South Korean cities
- · Transportation sector
- · Air quality management agencies
- · Machine learning researchers
- · Traditional forecasting methods
Improved decision-making for traffic and aviation in areas prone to low visibility.
Potential reduction in accidents and disruptions caused by poor atmospheric conditions.
Broader adoption of machine learning for localized environmental predictions across other critical atmospheric variables or geographies.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG