Skip to main content

Table 3 Confusion matrices for multilayer perceptron (MLP) and J48 using tenfold cross-validation, with columns signifying predictions and rows specifying actual cases

From: Machine learning and network analysis for diagnosis and prediction in disorders of consciousness

  1. The upper half (a) contains figures for the calculated MATADOC and CRS-R diagnostic outcomes using the 13 variables without and with primary cause of injury (PCOI) as a factor, and the lower half (b) for the three induced diagnostic categories without and with PCOI. The lead diagonals (in color) identify correct predictions