Research Article

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 6

DM.

Alternatives/criteriaPerformance evaluation metric criteria
Hybrid diagnosis modelsC1C2C3C4C5C6C7

A1ReF-decision treeC1-A1C2-A1C3-A1C4-A1C5-A1C6-A1C7-A1
A2ReF-SVMC1-A2C2-A2C3-A2C4-A2C5-A2C6-A2C7-A2
A3ReF-naive BayesC1-A3C2-A3C3-A3C4-A3C5-A3C6-A3C7-A3
A4ReF-KNNC1-A4C2-A4C3-A4C4-A4C5-A4C6-A4C7-A4
A5ReF-AdaBoostC1-A5C2-A5C3-A5C4-A5C5-A5C6-A5C7-A5
A6IG-decision treeC1-A6C2-A6C3-A6C4-A6C5-A6C6-A6C7-A6
A7IG-SVMC1-A7C2-A7C3-A7C4-A7C5-A7C6-A7C7-A7
A8IG-naive BayesC1-A8C2-A8C3-A8C4-A8C5-A8C6-A8C7-A8
A9IG-KNNC1-A9C2-A9C3-A9C4-A9C5-A9C6-A9C7-A9
A10IG-AdaBoostC1-A10C2-A10C3-A10C4-A10C5-A10C6-A10C7-A10
A11Chi2-decision treeC1-A11C2-A11C3-A11C4-A11C5-A11C6-A11C7-A11
A12Chi2-SVMC1-A12C2-A12C3-A12C4-A12C5-A12C6-A12C7-A12
A13Chi2-naive BayesC1-A13C2-A13C3-A13C4-A13C5-A13C6-A13C7-A13
A14Chi2-KNNC1-A14C2-A14C3-A14C4-A14C5-A14C6-A14C7-A14
A15Chi2-AdaBoostC1-A15C2-A15C3-A15C4-A15C5-A15C6-A15C7-A15

C: criteria; A: alternative; C1: train time; C2: test time; C3: AUC; C4: classification accuracy; C5: F1 score; C6: precision; C7: recall.