Feature Importance Assessment Results for Diabetes Prediction

Feature Importance Assessment Results is a small dataset of 5.5 KB created by Shucheng Feng and last updated on 2026-05-05. It contains results from an XGBoost feature importance evaluation used to screen key features for a diabetes prediction model. The dataset is shared on figshare under a CC-BY-4.0 license.

Use Cases

• Evaluating feature importance for diabetes prediction models based on the described XGBoost analysis.
• Benchmarking feature selection methods in clinical machine learning based on the described model construction process.
• Training or validating rule-based diagnostic models like EBRB based on the described streamlined rule base.

Strengths

• Dataset size is precisely 5.5 KB, indicating a focused and manageable scope.
• The dataset is associated with a specific methodological study proposing the NCR-EBRB model for diabetes diagnosis.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Shucheng Feng

Collection Method

Likely generated as part of a research study using XGBoost for feature importance evaluation.

Freshness

Last updated 2026-05-05 17:30:31; freshness should be verified.