Skip to main content

Table 5 Feature significance with respect to GAD7 using the permutation importance approach

From: Investigating the mental health of university students during the COVID-19 pandemic in a UK university: a machine learning approach using feature permutation importance

GAD7

Regression

KNN

SVM

Decision tree

Random forest

Gradient boost

Average

Gender

0.051

0.064

0.044

0.059

0.053

0.051

0.054

Ethnics

0.034

0.069

0.023

0.053

0.051

0.023

0.042

Education

0.019

0.071

0.027

0.060

0.058

0.033

0.045

Relationship

0.011

0.065

0.023

0.058

0.053

0.029

0.040

Adversity

0.080

0.049

0.077

0.051

0.068

0.094

0.070

Exercise

0.053

0.060

0.059

0.059

0.062

0.056

0.058

Alcohol

0.008

0.048

0.016

0.042

0.035

0.027

0.029

Tobacco

0.072

0.038

0.053

0.043

0.038

0.054

0.050

Relation-Impact

0.222

0.087

0.208

0.085

0.127

0.198

0.154

Communication

0.136

0.080

0.122

0.081

0.100

0.127

0.108

Therapy

0.146

0.065

0.126

0.092

0.072

0.090

0.099

Medication

0.049

0.060

0.063

0.072

0.061

0.043

0.058

Health-service

0.065

0.067

0.057

0.067

0.061

0.061

0.063

Social-distancing

0.008

0.025

0.013

0.030

0.023

0.024

0.020

Risk-group

0.012

0.043

0.025

0.051

0.047

0.032

0.035

Living-group

0.020

0.062

0.031

0.052

0.053

0.025

0.041

Contract-risk

0.015

0.047

0.032

0.045

0.041

0.034

0.036

  1. The top-5 most significant features highlighted in bold