Multi-class Diagnostic Performance Evaluation

Comprehensive evaluation of multi-class classification performance including per-class and overall metrics, ROC curves, confusion matrices, and model comparison capabilities.

Usage

multiclassdiagnostics(
  data,
  predicted,
  actual,
  positiveClass,
  confidenceLevel = 0.95,
  showROC = TRUE,
  showConfusion = TRUE,
  showPerClass = TRUE,
  showOverall = TRUE,
  compareModels = FALSE,
  predicted2,
  modelNames = "Model 1,Model 2",
  deLongTest = TRUE,
  mcnemarTest = TRUE,
  plotTheme = "default",
  saveResults = FALSE
)

Arguments

data

.

predicted

Variable containing predicted class labels

actual

Variable containing true class labels

positiveClass

For binary classification, specify which class is considered positive

confidenceLevel

Confidence level for intervals

showROC

Display ROC curves for each class (one-vs-rest)

showConfusion

Display confusion matrix

showPerClass

Show sensitivity, specificity, PPV, NPV for each class

showOverall

Show overall accuracy, kappa, and weighted metrics

compareModels

Enable model comparison

predicted2

Predicted classes for second model (for comparison)

modelNames

Names for the models being compared (comma-separated)

deLongTest

Perform DeLong test for ROC curve comparison (binary classification only)

mcnemarTest

Perform McNemar test for paired model comparison

plotTheme

Theme for plots

saveResults

Save detailed results to file

Value

A results object containing:

results$confusionMatrix					a table
results$perClassMetrics					a table
results$overallMetrics					a table
results$modelComparison					a table
results$deLongResults					a table
results$mcnemarResults					a table
results$rocPlot					an image
results$confusionPlot					an image
results$metricsPlot					an image
results$modelComparisonPlot					an image
results$text					a preformatted

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$confusionMatrix$asDF

as.data.frame(results$confusionMatrix)