From 4eb7e03fac4663228cf3938826b3c1508dd1c7cd Mon Sep 17 00:00:00 2001 From: Rohit Keswani Date: Mon, 14 Oct 2024 21:20:44 +0100 Subject: [PATCH] Replacing FPR to FNR in the FNR section --- book/3-classification.tex | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/book/3-classification.tex b/book/3-classification.tex index 86e614c..a54e85c 100644 --- a/book/3-classification.tex +++ b/book/3-classification.tex @@ -140,9 +140,9 @@ \subsection{False Negative Rate} % Left text with the image on the right \textbf{Figure 3.1 False Negative Rate.} \textbf{Top:} -3D surface illustrating FPR's non-linear relationship with FP and TN. FPR is lowest (blue) when FP is low. It increases (red) as FP increases. +3D surface illustrating FNR's non-linear relationship with FN and TP. FNR is lowest (blue) when FN is low. It increases (red) as FN increases. \textbf{Right:} -Shows how FPR decreases hyperbolically as total negative cases increase for fixed FP values. Lower FP maintains better FPR. +Shows how FNR decreases hyperbolically as total positive cases increase for fixed FN values. Lower FN maintains better FNR. \orangebox{% @@ -152,9 +152,9 @@ \subsection{False Negative Rate} This demonstrates the inherent trade-off between Type I and Type II errors in statistical testing. } -\textbf{FPR alternatives and related metrics} +\textbf{FNR alternatives and related metrics} -Other metrics used alongside or instead True Positive Rate (TPR/Recall/Sensitivity), Specificity, Precision, F1-Score, and ROC curve. +Other metrics used alongside or instead of False Negative Rate (TPR/Recall/Sensitivity), Specificity, Precision, F1-Score, and ROC curve. % ---------- FNR ---------- \clearpage