Fictitious Metabolic Health Data for Fatty Liver Disease Prediction

33.1 KB of synthetic data for machine learning education, containing demographic, anthropometric, clinical, and biochemical variables. The dataset was created by Himel Mondal and last updated on 2026-05-02. It is designed to support learning predictive modeling for fatty liver disease.

Use Cases

• Train a binary classifier to predict fatty liver disease based on anthropometric and clinical variables.
• Explore feature importance for metabolic health indicators like BMI, waist-to-hip ratio, and liver enzyme levels.
• Practice data preprocessing and feature engineering with a mix of continuous and categorical health variables.

Strengths

• Includes 17 clearly defined variables covering demographic, anthropometric, clinical, and biochemical domains.
• The dataset is small (33.1 KB), making it easy to download and process for educational purposes.
• Variables are directly relevant to a specific medical prediction task (fatty liver disease).

Limitations

• Row count is unknown, which may limit suitability assessment for certain modeling exercises.
• The dataset is explicitly fictitious, limiting its applicability to real-world research or clinical validation.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Synthetically generated for educational purposes.

Freshness

Last updated 2026-05-02 06:27:15; freshness should be verified.