Table 2 Degradation parameters estimated by machine learning methods: LDA, Naive bayes, Multi-class SVM.

From: Recognition of anxiety and depression using gait data recorded by the kinect sensor: a machine learning approach with data augmentation

Classification models

Hyperparameter

Value

LDA

Discrim type

‘Linear’

Gamma

0

Delta

0

Prior

Based on class frequency

Score transform

‘None’

Naive Bayes

Distribution names

‘Kernel’

Acquisition function name

Expected-improvement-plus

Verbose

0

Max objective evaluations

30

Multi-class SVM

Learners

‘Standardize’

Box constraint

Automatically searched

Kernel function

‘Polynomial’

Kernel scale

Automatically searched

Optimize hyperparameters

‘Kernel Function’

K-Fold

10

Max objective evaluations

30

Acquisition function name

‘Expected-improvement-plus’

Back to article page