F-Beta Ratio

Overview

Calculates the ratio of F-Beta scores between privileged and unprivileged groups.

Usage

Manually


# Calculate fb ratio

fbr = fb_ratio(df, protected_attribute, privileged_group, labels, positive_label, y_true, beta=1)
print(fbr)

Using Fairness Object

result = fo.compute(grouped_fb, beta=1)

Results

1.3749999996218751

These results are obtained by using the input data given in the Create Example Data page under Getting Started

Interpretation

A result of 1 indicates that the privileged and unprivileged groups have identical F-beta scores. A result > 1 indicates that privileged groups have better F-beta scores and a result < 1 indicates unprivileged groups have better F-beta scores.