New FI-TW Dataset Links Finnish Train Delays to Weather Data

Researchers have introduced FI-TW, a novel open dataset designed to analyze the impact of weather on railway delays in Finland. This dataset integrates operational data from the Digitraffic Railway Traffic Service with synchronized meteorological observations from 2018 to 2024, covering approximately 38.5 million observations across Finland's rail network. The dataset includes 28 engineered features and has been preprocessed to handle missing data and outliers. Initial analysis using XGBoost regression demonstrated the dataset's utility by achieving a Mean Absolute Error of 2.73 minutes in predicting station-specific delays, highlighting its potential for machine learning applications in railway operations. AI

IMPACT Enables new machine learning applications for predicting train delays and assessing weather impacts on railway infrastructure.

RANK_REASON The cluster describes a new research dataset and its application in a scientific paper. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Digitraffic Railway Traffic Service
• Finland
• FI-TW
• Gotit.pub
• Hugging Face
• Vinicius Pozzobon Borin
• XGBoost

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →