Feature Signature Discovery for Autism Detection: An Automated Machine Learning Based Feature Ranking Framework

Table 5

Comparative study on the performance of AutoML models with previous work.

Classification model	Dataset—classifier accuracy (%)
	Toddler	Child	Adolescent	Adult
Neural network [24]	99
C4.5 [24]	96
Deep neural network [23]		92
SGD (stochastic gradient descent) [7]		99.6
Random forest [9]			97.2
Soft voting classifier [8]				94.45
Proposed AutoML—all features	96.7	97.6	94.5	96.8
Proposed AutoML—Feature selection	99.6	100	94.5	98.1
Proposed aggressive feature selective classifier	95.8	100	87	98.1