You are on page 1of 5

BUSINESS ANALYTICS PROJECT

TOPIC: CONFUSION MATRIX

SUBMITTED BY SUDESHNA PAUL


REGISTRATION NO. -19PGPM137

SUBMITTED TO DR. KRR GANDHI


PROBLEM: Let’s take the case of the Beta Thalassemia
patients classification problem.

The goal is to classify Thalassemia patients in two classes: major vs.


minor (carrier).

My case contains 100 Beta Thalassemia patients described through


various symptoms,
Now we need to answer several questions:

● How many of the actual minor were predicted as major?


● How many major were predicted as minor?
● How many of the actual minor were predicted as minor?
● How many major were predicted correctly?

CONFUSION MATRIX

FORMULAS:

SENSITIVITY= (TRUE POSITIVE/ACTUAL YES)

SPECIFICITY= (TRUE NAGATIVE/ACTUAL NO)

FALSE POSITIVE RATE= (FALSE POSITIVE/ACTUAL NO)

FALSE NAGATIVE RATE= (FALSE NEGATIVE/ACTUAL NO)

ACCURACY= (TRUE POSITIVE+TRUE NAGATIVE)/TOTAL


ERROR RATE= (FALSE POSITIVE+FALSE NEGATIVE)/TOTAL

PRECISION= (TRUE POSITIVE/PREDICTED YES)

PREVALENCE= (ACTUAL YES/TOTAL)

In our case,

Beta Thalassemia minor is acceptable as healthy, so positive.

Beta Thalassemia major is not acceptable as healthy, so negative.

Confusion matrix explained


● The cell identified by the row and column for the positive class
contains the True Positives, i.e. where the actual and predicted class
is minor.
● Cells identified by rows for the negative class and the column for the
positive class contain the False Positives, where the actual class is
major, and the predicted class is minor.
● Cells identified by the row for the positive class and columns for the
negative class contains the False Negatives, where the actual class is
minor, and the predicted class is major.
● Cells identified the row and column for the negative class contain
the True Negatives, where the actual class is major, and the
predicted class is also major.

Summary

So we have seen that confusion matrix shows the performance of a


classification model: how many positive and negative events are
predicted correctly or incorrectly. It helps to take decisions from the
data.

You might also like