Table 4 Root-mean square error (RMSE) loss for independent variables of best performing ML models, including the direction of association, to detect primary cluster ZIP5 neighborhoods for each oral disease. Values of variables over 0.01 RSME loss are shown in bold
From: Detecting high-risk neighborhoods and socioeconomic determinants for common oral diseases in Germany
Variable | Periodontitis (SVM) | Caries (severe) (SVM) | Irreversible Pulpitis (NN) | Tooth Loss (LR) |
|---|---|---|---|---|
Baseline | 0.059 | 0.051 | 0.092 | 0.013 |
Education | 0.004 | 0.007 | 0.011 | 0.002 |
Inhabitants per household | 0.003 | −0.003 | −0.022 | −0.019 |
Inhabitants per km2 | −0.020 | −0.023 | −0.012 | 0.011 |
Incidence other oral diseases | −0.005 | 0.000 | −0.001 | 0.003 |
Income | −0.011 | −0.016 | −0.004 | −0.010 |
Occupation | −0.001 | 0.004 | −0.008 | −0.006 |
Conservative upscale milieu | 0.009 | 0.009 | 0.000 | 0.005 |
Nostalgic middle-class milieu | 0.004 | 0.003 | 0.039 | 0.009 |
Precarious milieu | 0.004 | 0.003 | 0.002 | 0.018 |
Traditional milieu | 0.019 | 0.017 | 0.008 | −0.004 |
Dentists per 1000 inhabitants | −0.001 | − 0.001 | 0.000 | 0.003 |