Benchmark of 80 Machine Learning Models for Cardiovascular Risk Prediction

80 machine learning models were benchmarked for cardiovascular risk prediction using the Kaggle Cardiovascular Disease dataset. The results, ordered by AUC-ROC, compare four base architectures across 20 distinct methodological experiments. The dataset was published by 'Predicción Cardiovascular' on figshare in June 2026.

Use Cases

• Compare the performance of XGBoost, LightGBM, Random Forest, and Logistic Regression based on the benchmark results.
• Analyze the trade-off between model accuracy and hyperparameter optimization time based on the 'Tiempo Búsqueda (s)' and 'AUC-ROC' metrics.
• Study the impact of different data preprocessing and feature engineering strategies based on the 20 distinct experimental designs.
• Benchmark new cardiovascular risk models against the reported AUC-ROC and F1-Score results.

Strengths

• Contains results for 80 distinct machine learning models, providing a broad comparison.
• Benchmarks four major algorithm families (XGBoost, LightGBM, Random Forest, Logistic Regression) across 20 experimental designs.
• Primary performance metric (AUC-ROC) is clearly defined and used for ranking, with a maximum reported value of 0.8059.
• Includes computational efficiency metrics like hyperparameter optimization time ('Tiempo Búsqueda (s)').

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The underlying clinical data features are not described, only the model performance metrics.

Provenance

Source

figshare, author 'Predicción Cardiovascular'.

Collection Method

Results from an experimental evaluation of models trained on the Kaggle Cardiovascular Disease dataset.

Freshness

Last updated 2026-06-04 17:22:14; freshness should be verified.