Skip to main content

Table 4 Model performance of models built with different machine learning algorithms

From: Predicting later categories of upper limb activity from earlier clinical assessments following stroke: an exploratory analysis

 

Model name

Model Performance Statistic

Single Decision Tree

Small

Medium

Large

Bagged model

Random forest

Bagged model

Random forest

Bagged model

Random forests

In-sample

       

Accuracy*

       

 Mean

0.70

0.96

0.96

1.0

1.0

1.0

1.0

 IQR

(0.56, 0.82)

(0.87, 1.0)

(0.87, 0.99)

(0.93, 1.0)

(0.93, 1.0)

(0.93, 1.0)

(0.93, 1.0)

Out-of-bag estimate of error^

       

 Mean

95% CI

na

0.55

(0.54,0.56)

0.52

(0.52,0.54

0.47

(0.46, 0.48)

0.46

(0.44, 0.46)

0.48

(0.46, 0.48)

0.48

(0.48,0.48)

  1. *In-sample accuracy is a measure of the explanatory power of the model and was quantified for all seven models by comparing the predicted category and the actual UL performance categories. Values closer to 1.00 indicate better model performance
  2. ^Out-of-bag estimate of error is a measure of the predictive power of the models and was quantified for the six bagging models as cross-validation accuracy. Lower error-rate values indicate better model performance
  3. CI: confidence interval; IQR: inter-quartile range