Sale!

# Binary Classification and Performance Measures

\$30.00

Category:

Binary Classification and Performance Measures
Active Learning:
Suppose that Fereydoon has developed a binary classification model to classify Autism Spectrum Disorder (ASD)
based on fMRI data. The model has been trained on a large pool of heterogeneous multi-site datasets with subjects
aged 1-55 year-old, and with balanced number of the two classes (almost equal number of ASD-Positive vs ASDNegative).
The following table shows the results of testing the model on a small dataset that contains fMRI data from only
one site, recorded from 10 subjects (age range 2-18 year-old). Actual classes have been confirmed by expert
psychiatrists.
Subject ID Actual Class Predicted Class
S1 ASD-Positive ASD-Positive
S2 ASD-Positive ASD-Positive
S3 ASD-Positive ASD-Positive
S4 ASD-Negative ASD-Negative
S5 ASD-Negative ASD-Positive
S6 ASD-Negative ASD-Positive
S7 ASD-Negative ASD-Positive
S8 ASD-Negative ASD-Negative
S9 ASD-Positive ASD-Positive
S10 ASD-Negative ASD-Negative

(A) Count the following from the table:
➢ True Positive (TP): Number of rows where “Actual Class” is ASD-Positive and “Predicted Class” is also
ASD-Positive.
➢ True Negative (TN): Number of rows where “Actual Class” is ASD-Negative and “Predicted Class” is also
ASD-Negative.
➢ False Positive (FP): Number of rows where “Actual Class” is ASD-Negative but “Predicted Class” is ASDPositive.
➢ False Negative (FN): Number of rows where “Actual Class” is ASD-Positive but “Predicted Class” is ASDNegative.
Binary Classification and Performance Measures
(B) Based on the numbers you just calculated in part (A), fill in the confusion matrix:
TP = FP =
FN = TN =
(C) What is the accuracy of this classifier? Accuracy = (TN + TP)/(TN+TP+FN+FP)
(D) What is the Precision and Recall of this classifier? Precision = TP/(TP+FP)
Recall = TP/(TP+FN)
(E) What is the F1 Score of this classifier? F1 Score = TP/(TP+((FP+FN)/2))

Binary Classification and Performance Measures
\$30.00
Hello
Can we help?