Table 13 Classification results of datasets after feature selection under RF classifier
From: Missing data imputation, prediction, and feature selection in diagnosis of vaginal prolapse
Imputation methods | FSM(No. of selected features) | Accuracy | F1 | AUC |
|---|---|---|---|---|
Mean | LASSO(20) | 0.7995(0.0105) | 0.8541(0.0085) | 0.7659(0.0119) |
SCAD(41) | 0.7945(0.0109) | 0.8499(0.0091) | 0.7617(0.0122) | |
BAR(7) | 0.7303(0.0136) | 0.8163(0.0102) | 0.6451(0.0172) | |
EM | LASSO(20) | 0.7558(0.0114) | 0.8308(0.0087) | 0.6853(0.0133) |
SCAD(30) | 0.7645(0.0124) | 0.8359(0.0098) | 0.6988(0.0137) | |
BAR(7) | 0.7289(0.0123) | 0.8134(0.0098) | 0.6500(0.0138) | |
KNN | LASSO(20) | 0.7677(0.0111) | 0.8375(0.0088) | 0.7045(0.0129) |
SCAD(38) | 0.7907(0.0118) | 0.8485(0.0095) | 0.7523(0.0135) | |
BAR(7) | 0.7306(0.0106) | 0.8150(0.0086) | 0.6506(0.0124) | |
DAE | LASSO(27) | 0.7018(0.0114) | 0.7896(0.0095) | 0.6339(0.0125) |
SCAD(62) | 0.7009(0.0112) | 0.7875(0.0091) | 0.6372(0.0134) | |
BAR(7) | 0.6895(0.0135) | 0.7810(0.0108) | 0.6193(0.0145) | |
GAIN | LASSO(10) | 0.8011(0.0076) | 0.8501(0.0092) | 0.7680(0.0127) |
SCAD(52) | 0.8025(0.0099) | 0.8559(0.0081) | 0.7692(0.0118) | |
BAR(9) | 0.7986(0.0061) | 0.8508(0.0051) | 0.7705(0.0078) |