Fig 1.
Flowchart for the proposed methodology.
Table 1.
Creation of target variable. As stepwise procedure of how the target variable was created.
Fig 2.
The population at risk of NTDs by region.
Table 2.
Frequency distribution of co-endemicity classes.
Fig 3.
Distribution of STH endemicity levels.
Fig 4.
Distribution of SCH endemicity levels.
Fig 5.
Distribution of LF endemicity levels.
Fig 6.
NTDs co-endemicity status by region.
Fig 7.
MDA strategies and usage by region.
Table 3.
Correlation matrix. The correlation matrix shows the relationship between the target variable and the predictors.
Fig 8.
STH prevalence by region in Kenya.
Table 4.
Model overall performance metrics and statistics of the RF model.
Table 5.
RF model key performance metrics.
Fig 9.
Random Forest plot.
Fig 10.
RF Variable importance.
Fig 11.
RF ROC AUC.
Table 6.
Resampling results across tuning parameters for the GBM model.
Fig 12.
GBM model hyper parameter results.
Fig 13.
GBM variable importance.
Fig 14.
ROC curve of the GBM model.
Table 7.
Performance metrics of the GBM model.
Table 8.
XGBoost performance metrics.
Fig 15.
XGBoost model illustration.
Fig 16.
XGBoost ROC curve.
Fig 17.
XGBoost variables importance.
Fig 18.
Models comparison.
Table 9.
Models performance based on the ROC curves.
Table 10.
Models performance based on selected metrics.