Fig. 1: Comparison of the top 25% of features ranked by aggregate feature importance across all pairwise models with the 3-degree SVM polynomial kernel. | Translational Psychiatry

Fig. 1: Comparison of the top 25% of features ranked by aggregate feature importance across all pairwise models with the 3-degree SVM polynomial kernel.

From: Relative importance of speech and voice features in the classification of schizophrenia and depression

Fig. 1: Comparison of the top 25% of features ranked by aggregate feature importance across all pairwise models with the 3-degree SVM polynomial kernel.

Feature importance was computed as the post hoc permutation feature importance. For each model, there were twenty participants per group (HC healthy controls, SSD schizophrenia spectrum disorder, MDD major depressive disorder) and four speech samples per participant.

Back to article page