Development of Hepatitis Disease Detection System by Exploiting Sparsity in Linear Support Vector Machine to Improve Strength of AdaBoost Ensemble Model
Table 4
Comparison of the proposed method with some well-known methods proposed for hepatitis disease in terms of prediction accuracy [25, 28, 38, 39].
Model or method number
Model or method
Study or authors
Acc. (%)
1
K-nearest neighbours (KNN)
Nilashi et al.
71.41
2
Neural network
Nilashi et al.
78.31
3
ANaFIS
Nilashi et al.
79.67
4
SVM
Nilashi et al.
81.17
5
ASI
Stern and Dobnikar
82.0
6
Multilayer perceptron + backpropagation
Adamczak
77.4
7
Linear discriminant analysis (LDA)
Stern and Dobnikar
86.4
8
Multilayer perceptron (MLP)
Ozyildirim, yildirim
74.37
9
Radial basis function (Tooldiag)
Adamczak
79.0
10
1NN
Stern and Dobnikar
85.3
11
Radial basis function (RBF)
Ozyildirim, yildirim
83.75
12
15NN, stand. Euclidean
Grudzinski
89.0
13
FSM with rotations
Adamczak
89.7
14
FSM without rotations
Adamczak
88.5
15
Multilayer perceptron with backpropagation
Stern and Dobnikar
82.1
16
Quadratic discriminant analysis
Stern and Dobnikar
85.8
17
(NB and semi-NB), i.e., Naive Bayes and semi-NB
Stern and Dobnikar
86.3
18
Fisher discriminant analysis (FDA)
Stern and Dobnikar
84.5
19
LVQ
Stern and Dobnikar
83.2
20
GRNN
Ozyildirim, yildirim
80.0
21
ASR
Stern and Dobnikar
85.0
22
IncNet
Norbert jankowski
86.0
23
Classification and regression tree (decision tree)
