Table 7 Predictors, extracted from PDA data set, ordered according to their significance for each dimensionality reduction method.

Ord	Decision tree	Random forest	Gradient boosting	Logistic regression	LASSO	Expert selection (PDA_EXP)
Ord	Decision tree	Random forest	Gradient boosting	Logistic regression	LASSO	Initial	Num	Avg	Final
1	Size_LD	Size_LD	Koos_LD	Size_LD	Koos_LD	Koos_LD	5	2	Koos_LD
2	Koos_LD	Koos_LD	Size_SC	PTA_VSAR4_SC	PTA_DAR4_IC	Size_LD	4	1.5	-
3	Size_AC	Size_SC	Size_LD	Size_SC	SRT_IC	Size_SC	4	4.5	Size_SC
4	PTA_VSAR4_AC	SRT_LD	PTA_VSSC4	Koos_LD	SRT_LD	Koos_TD	3	6.7	-
5	Koos_TD	Koos_TD	SRT_SC	PTA_DAR8_SC	PTA_VSAR8_IC	PTA_DAR4_SC	3	7.7	PTA_DAR4_SC
6	PTA_VSAR8_AC	PTA_VSAR8_IC	PTA_VSAR8_SC	PTA_VSAR8_SC	PTA_DAR8_LD	PTA_VSAR4_SC	3	5.3	PTA_VSAR4_SC
7	PTA_DAR8_IC	PTA_DAR4_SC	PTA_VSAR4_LD	PTA_VSAR8_SC	PTA_DAR8_IC	PTA_DAR8_IC	3	7.7	–
8	PTA_DAR4_SC	PTA_DAR4_LD	PTA_DAR8_SC	PTA_DAR4_SC	PTA_VSAR4_IC	Size_AC	3	7.3	–
9	SRT_AC	PTA_DAR8_AC	PTA_DAR8_LD	PTA_DAR8_IC	Size_AC	–	–	–	–
10	Size_SC	PTA_VSAR4_SC	Size_AC	Koos_TD	PTA_VSAR4_AC	–	–	–	–

Irrelevant variables were rejected from expert selection. Variables and metrics are expressed as follows (see “Methods”): (a) suffix 4 holds for the basic frequency range, suffix 8 holds for the full frequency range; (b) subscripts H and VS hold for the healthy or diseased ear, respectively, subscript D holds for the difference of averaged PTA values between the two ears; (c) tailing abbreviations have the following meaning: AR average row wise, IR intercept row wise, SR slope row wise, AC average column wise, IC intercept column wise, SC slope column wise, LD last difference, TD total difference.

Quick links

Search