Table 5 Detailed performance analysis of prediction model without random under-sampling.

From: A Systematic Machine Learning Based Approach for the Diagnosis of Non-Alcoholic Fatty Liver Disease Risk and Progression

 

Class

Desirable

Borderline_high

High

Very_High

Weighted Avg.

TP Rate

0.937

0.05

0.296

0.024

0.762

FP Rate

0.777

0.03

0.053

0.001

0.542

Precisionµ

0.78

0.493

0.573

0.133

0.669

Recallµ

0.937

0.451

0.396

0.024

0.735

F-Measureµ

0.851

0.279

0.349

0.04

0.676

PrecisionM

0.757

0.561

0.651

0.416

0.677

RecallM

0.832

0.59

0.503

0.366

0.713

MCC

0.328

0.247

0.195

0.055

0.299

AROC

0.748

0.631

0.738

0.507

0.731

Back to article page