Insurance Company Benchmark Data from COIL 2000

COIL 2000 benchmark data provides a standardized set for insurance analytics. It was contributed to the UCI Machine Learning Repository for a competition. The dataset is used to model customer behavior and insurance product attributes.

Use Cases

• Predict customer insurance product purchase likelihood using demographic and policyholder features.
• Analyze risk factors and customer segmentation based on attributes like vehicle type or region.
• Benchmark binary classification models on the target variable indicating policy subscription.
• Perform feature importance analysis to identify key drivers in insurance marketing campaigns.

Strengths

• Standardized benchmark from a known machine learning repository (UCI).
• Designed for a real-world predictive modeling task in the insurance domain.

Limitations

• Specific column names, data types, and row counts are not provided in the input.
• The temporal coverage and geographic scope of the original data are unknown.

Provenance

Source

UCI Machine Learning Repository.

Collection Method

Originally compiled for the COIL 2000 data mining competition.