Hybrid Diagnosis Models for Autism Patients Based on Medical and Sociodemographic Features Using Machine Learning and Multicriteria Decision-Making (MCDM) Techniques: An Evaluation and Benchmarking Framework

Table 10

FDOSM results of the benchmarking of the 15 hybrid diagnosis models.

Hybrid diagnosis models		Score value	Ranking order
A1	ReF-decision tree	0.157142829	1
A2	ReF-SVM	0.761111002	10
A3	ReF-naive Bayes	0.576984072	7
A4	ReF-KNN	0.781745832	13
A5	ReF-AdaBoost	0.350793653	4
A6	IG-decision tree	0.175396799	2
A7	IG-SVM	0.761111002	10
A8	IG-naive Bayes	0.6007936	8
A9	IG-KNN	0.797618813	14
A10	IG-AdaBoost	0.371428574	5
A11	Chi^2-decision tree	0.294444441	3
A12	Chi^2-SVM	0.776983983	12
A13	Chi^2-naive Bayes	0.6007936	8
A14	Chi^2-KNN	0.816666435	15
A15	Chi^2-AdaBoost	0.400000007	6