Skip to main content
Advertisement

< Back to Article

Fig 1.

Flowchart for the proposed methodology.

More »

Fig 1 Expand

Table 1.

Creation of target variable. As stepwise procedure of how the target variable was created.

More »

Table 1 Expand

Fig 2.

The population at risk of NTDs by region.

More »

Fig 2 Expand

Table 2.

Frequency distribution of co-endemicity classes.

More »

Table 2 Expand

Fig 3.

Distribution of STH endemicity levels.

More »

Fig 3 Expand

Fig 4.

Distribution of SCH endemicity levels.

More »

Fig 4 Expand

Fig 5.

Distribution of LF endemicity levels.

More »

Fig 5 Expand

Fig 6.

NTDs co-endemicity status by region.

More »

Fig 6 Expand

Fig 7.

MDA strategies and usage by region.

More »

Fig 7 Expand

Table 3.

Correlation matrix. The correlation matrix shows the relationship between the target variable and the predictors.

More »

Table 3 Expand

Fig 8.

STH prevalence by region in Kenya.

More »

Fig 8 Expand

Table 4.

Model overall performance metrics and statistics of the RF model.

More »

Table 4 Expand

Table 5.

RF model key performance metrics.

More »

Table 5 Expand

Fig 9.

Random Forest plot.

More »

Fig 9 Expand

Fig 10.

RF Variable importance.

More »

Fig 10 Expand

Fig 11.

RF ROC AUC.

More »

Fig 11 Expand

Table 6.

Resampling results across tuning parameters for the GBM model.

More »

Table 6 Expand

Fig 12.

GBM model hyper parameter results.

More »

Fig 12 Expand

Fig 13.

GBM variable importance.

More »

Fig 13 Expand

Fig 14.

ROC curve of the GBM model.

More »

Fig 14 Expand

Table 7.

Performance metrics of the GBM model.

More »

Table 7 Expand

Table 8.

XGBoost performance metrics.

More »

Table 8 Expand

Fig 15.

XGBoost model illustration.

More »

Fig 15 Expand

Fig 16.

XGBoost ROC curve.

More »

Fig 16 Expand

Fig 17.

XGBoost variables importance.

More »

Fig 17 Expand

Fig 18.

Models comparison.

More »

Fig 18 Expand

Table 9.

Models performance based on the ROC curves.

More »

Table 9 Expand

Table 10.

Models performance based on selected metrics.

More »

Table 10 Expand