Scribd
Upload
53 views2 pages

5 ROC Curve

The receiver operating characteristic curve (ROC) is a visual tool used to compare classification models by plotting their true positive rate against their false positive rate. It shows the tradeoff between sensitivity and specificity for a binary classifier. The area under the ROC curve is a measure of a model's accuracy, with an area of 1 representing a perfect classifier and an area of 0.5 or less representing a failure. The document analyzes the ROC curves of four algorithms applied to sample data, finding that two algorithms, CGF and CGP111, produced the most accurate results as evidenced by their higher ROC areas.

Uploaded by

Archana
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
53 views2 pages

5 ROC Curve

The receiver operating characteristic curve (ROC) is a visual tool used to compare classification models by plotting their true positive rate against their false positive rate. It shows the tradeoff between sensitivity and specificity for a binary classifier. The area under the ROC curve is a measure of a model's accuracy, with an area of 1 representing a perfect classifier and an area of 0.5 or less representing a failure. The document analyzes the ROC curves of four algorithms applied to sample data, finding that two algorithms, CGF and CGP111, produced the most accurate results as evidenced by their higher ROC areas.

Uploaded by

Archana
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Receiver operating characteristic curve

S. Sumitra
The receiver operating characteristic curve (ROC) of a binary classifier consists
of points with (1- specificity) (FPR = False Positive Rate) as the x co-ordinate and
sensitivity (TPR = True positive Rate) as the y co-ordinate (see figure (1)). It is a
visual tool for comparing classification models. It shows a tradeoff between sensitivity
and specificity. The points in the curve are obtained using different threshold values
for classification.
A completely random guess would result a point in the ROC curve along a diagonal line from the left bottom to the top right corner. Points below the diagonal
are worse than random guessing. The accuracy of the test is higher if the ROC plot
bulges towards the upper left corner.
[The point (0,1) is the perfect classifier as it classifies all positive cases and negative cases correctly. The point (0,0) represents a classifier that predicts all cases to
be negative, while the point (1,1) corresponds to a classifier that predicts every case
to be positive. Point (1,0) is the classifier that is incorrect for all classifications.
In many cases, a classifier has a parameter that can be adjusted to increase TP at
the cost of an increased FP or decrease FP at the cost of a decrease in TP. Each
parameter setting provides a (FP, TP) pair and a series of such pairs can be used to
plot an ROC curve.]
The value of the area under ROC curve lies in the interval [0,1] and it is a measure
of the accuracy of the model. An area of 1 represents a perfect test while an area less
than or equal to 0.5 represents a total failure. More area under the curve signifies that
it is identifying more true positives while minimizing the number of false positives.
The figure (1) shows the ROC curve obtained by applying four different algorithms on the data under study. From the figure (1), it is clear that the algorithms
CGF and CGP111 produce more accurate classification results than the other two
algorithms, as they have the highest ROC area among the four algorithms.

1
0.9
CGF ROC Area= 0.9403

0.8

CGPI ROC Area= 0.9328

Sensitivity

0.7

CGPII ROC Area= 0.9402


CGPIII ROC Area= 0.9403

0.6
0.5
0.4
0.3
0.2
0.1
0

0.2

0.4

0.6

0.8

(1Specificity)

Figure 1: ROC curve and ROC area.

You might also like