# Load the data and fit the logistic regression model

breast read.csv $($file$.$choose$(""),$ header $=$ TRUE$)$

attach $($breast$)$

model glm$($as$.$factor$($BenignMalignant$)$ ClumpThickness $+$ CellsizeUniformity $+$ CellshapeUniformity $+$

binomial$($link $=$ "logit"$))$

# Make predictions

breastPrediction predict$($model$2,$ breast, type $=$ "response"$)$

# Convert predicted probabilities to class labels

breastPrediction$2\_$class ifelse$($breastPrediction$2>0\mathrm{.}5,4,2)$

# Create a contingency table for actual vs$.$ predicted classes

confusion$\_$matrix table$($as$.$factor$($breast$BenignMalignant$),$ as$.$factor$($breastPrediction$2\_$class$))$

# Compute performance metrics

confusion matrix $2,2$

confusion$\_$matrix$[1,1]$

FP confusion$\_$matrix$[1,2]$

FN confusion$\_$matrix $2,1$

FNR FN $\frac{?}{P}$

TNR

# Display the computed performance metrics

cat $("$Positive $($P$)$:$",$ P$,"$

$")$

cat $("$Negative $($N$)$:$",$ N$,"$

$")$

cat$("$True positive $($TP$)$:$",$ TP$,"$

$")$

cat$("$True negative $($TN$)$:$",$ TN$,"$

$")$

cat$("$False positive $($FP$)$:$",$ FP$,"$

$")$

cat$("$False negative $($FN$)$:$",$ FN$,"$

$")$

cat$("$True positive rate $($TPR$)$:$",$ TPR$,"$

$")$

cat$("$False negative rate $($FNR$)$:$",$ FNR$,"$

$")$

cat$("$False positive rate $($FPR$)$:$",$ FPR$,"$

$")$

cat$("$True negative rate $($TNR$)$:$",$ TNR$,"$

$")$