Algorithms
No. of attributes
Time elapsed in second
Accuracy in %
WTPR
WFPR
WROC
J48 Decision tree
28
0.48
95.90
96.5
3.5
0.998
26
0.45
95.86
96.4
3.7
0.996
24
0.38
95.87
96.3
3.7
0.989
22
0.37
95.87
96.3
3.7
0.989
20
0.34
95.80
96.3
3.7
0.989
16
0.23
94.77
94.7
5.3
0.99
Random tree decision tree
28
0.08
95.75
96.2
3.9
0.980
26
0.05
95.90
96.3
3.8
0.982
24
0.09
95.59
96.2
3.9
0.981
22
0.11
95.80
96.2
3.8
0.980
20
0.13
95.45
96.0
4.0
0.981
16
0.05
94.75
94.8
5.3
0.98|1
REP tree decision tree
28
0.27
95.90
96.3
3.7
0.990
26
0.42
95.86
96.3
3.7
0.989
24
0.22
95.84
96.2
3.8
0.987
22
0.22
95.84
96.2
3.8
0.988
20
0.36
95.58
96.1
3.9
0.988
16
0.13
94.72
94.3
5.7
0.983
Multilayer perceptron ANN
28
2567.88
94.82
94.8
5.2
0.977
26
3797.87
95.13
95.1
4.9
0.975
24
1418.18
95.05
95.0
5.0
0.977
22
1342.15
95.07
95.1
5.0
0.976
20
950.23
94.99
95.0
5.0
0.980
16
858.6
94.60
94.6
5.4
0.979
Table 12:
Comparison of performances on J48, Random tree, REP tree and multilayer perceptron algorithms to predict under-five children admission to pediatric ward in NEMM Hospital, 2012.
