Skip to main content
Fig. 1 | BMC Public Health

Fig. 1

From: Prediction of metabolic and pre-metabolic syndromes using machine learning models with anthropometric, lifestyle, and biochemical factors from a middle-aged population in Korea

Fig. 1

Feature importance in the MetS prediction model. a Feature importance when using 12 features; (b) Feature importance when using 20 features. Variable importance results when building the model are presented. BMI, body mass index; WHR, waist-to-hip ratio; PA, physical activity; KM type, Korean medicine type; HOMA-IR, homeostatic model assessment for insulin resistance; GGT, gamma-glutamyl transferase; HbAlc, hemoglobin A1c; hsCRP, high sensitivity C-reactive protein; ALT, alanine transaminase; ALP, alkaline phosphatase; AST, aspartate transaminase

Back to article page

BMC Public Health

ISSN: 1471-2458

Contact us