Oblique decision trees for spatial pattern detection: optimal algorithm and application to malaria risk

Table 1 Spatial pattern resulting from the ODT-model. The first line refers to the areas without any partition.

	Centroid's Coordinates^a	Pop.^b	Risk of infection [CI95%]	Number of Locations^c
No pattern	X = -8.266497256 Y = 12.20520982	1 461	32.44% [30.09–34.89]	164
P1	X = -8.270634 Y = 12.202594	30	26.67% [14.18–44.45]	5
P2	X = -8.27019 Y = 12.20438615	153	60.13% [52.22–67.55]	13
P3	X = -8.26849 Y = 12.1999733	26	50.0% [32.06–67.94]	3
P4	X = -8,2659751 Y = 12.205486	1 252	28.83% [26.39–31.4]	143

a- The coordinates are for the centroid of each partition.
b- Pop. refers to the total number of children included in each partition.
c- The number of locations refers to the total number of households within each partition.

ISSN: 1471-2288