Skip to main content
. 2022 Feb 3;7(1):262–276. doi: 10.1016/j.idm.2022.01.004

Fig. 10.

Fig. 10

Summary plot of the complete dataset. The horizontal axis represents the SHAP values of each of the variables by which the model predicts the number of hookworm infected individuals per household while the vertical axis lists all of the variables used in the study. The y-axis represents the variables used in the study, which refer to: Electricit = Presence or absence of electricity; ri_orcd_n = Constructed risk from soil organic carbon content; BSI_n = Bare Soil Index; Clyp_n = Weight percentage of the clay particles; MSI_n = Moisture Stress Index; Phikcl_n = pH index measured in KCl solution; Ocstha_n = Soil organic carbon stock; BLDFIE_n = Bulk density; NDBI_n = Normalized Difference Builtup; Drinking_1 = Source of drinking water during the rainy season; Drinking w = Source of drinking water during drought periods; Phihox_n = pH index measured in water solution; Water cook = Source of water used for cooking; Hand washi = Hand washing, source of water used for hand washing; Domestic animals = Presence or absence of domestic animals; SLTppt_n = Weight percentage of the silt particles; Radio = Presence or absence of radio; Toilet = Type of bath; ri_bfd = Constructed risk from bulk density; Bath = Origin of water used in the bath. The x-axis represents the number of hookworm infected individuals per household which is the output value of the model.