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

Brain Informatics

Table 9 Feature significance with respect to Support-needs using the permutation importance approach

Support-needs	Regression	KNN	SVM	Decision tree	Random forest	Gradient boost	Average
Gender	0.019	0.059	0.024	0.053	0.051	0.037	0.040
Ethnics	0.102	0.066	0.082	0.064	0.063	0.094	0.078
Education	0.016	0.069	0.013	0.063	0.066	0.031	0.043
Relationship	0.009	0.063	0.015	0.060	0.063	0.024	0.039
Adversity	0.109	0.045	0.097	0.064	0.069	0.120	0.084
Exercise	0.044	0.051	0.058	0.063	0.060	0.042	0.053
Alcohol	0.016	0.044	0.013	0.048	0.044	0.034	0.033
Tobacco	0.014	0.040	0.015	0.037	0.031	0.019	0.026
Relation-Impact	0.090	0.064	0.128	0.070	0.068	0.079	0.083
Communication	0.051	0.063	0.052	0.064	0.066	0.046	0.057
Therapy	0.051	0.076	0.113	0.064	0.058	0.059	0.070
Medication	0.113	0.073	0.098	0.075	0.071	0.088	0.086
Health-service	0.186	0.088	0.183	0.090	0.102	0.149	0.133
Social-distancing	0.016	0.030	0.011	0.026	0.021	0.025	0.022
Risk-group	0.058	0.047	0.023	0.055	0.047	0.064	0.049
Living-group	0.021	0.064	0.027	0.052	0.058	0.023	0.041
Contract-risk	0.084	0.056	0.048	0.052	0.062	0.068	0.062