Credit Risk Assessment Data for 594 Private Long-Term Care Institutions in Hangzhou, China

594 private long-term care institutions in Hangzhou, China form the basis for an interpretable machine-learning framework for credit risk assessment. The dataset likely contains institution-level features used to predict creditworthiness, including registered capital, tax-paying employees, financing history, and patent counts. The framework was developed by Zhouyi Gu and published on figshare in April 2026.

Use Cases

• Training credit risk prediction models based on institution-level financial and operational features.
• Identifying key predictors of institutional creditworthiness using feature importance analysis.
• Deploying a real-time credit rating tool for families and regulators based on the trained model.
• Comparing the performance of different machine-learning algorithms with class-imbalance handling techniques.

Strengths

• Contains data from 594 specific institutions, providing a concrete sample size.
• Focuses on objective, publicly available data for credit assessment, as described.
• The final model is deployed in a web application, suggesting practical applicability.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is 108.0 KB, indicating a very limited scope and likely small number of features.

Provenance

Source

Publicly available records from private long-term care institutions in Hangzhou, China.

Collection Method

Data was gathered and used to develop and validate a machine-learning framework.

Freshness

Last updated 2026-04-29 05:37:14; freshness should be verified.

Geography

Hangzhou, China