Imbalanced Data and Algorithmic Fairness

0.0(0)
studied byStudied by 0 people
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
Card Sorting

1/18

flashcard set

Earn XP

Description and Tags

Flashcards about Imbalanced Data and Algorithmic Fairness

Study Analytics
Name
Mastery
Learn
Test
Matching
Spaced

No study sessions yet.

19 Terms

1
New cards

overall error

Classifiers can be biased towards the majority class because they try to reduce the .

2
New cards

Oversampling, Undersampling, Create synthetic data

Three methods of Data Balancing are: __, __, and __.

3
New cards

oversampling

Random __: randomly duplicate data points from the minority class.

4
New cards

undersampling

Random __: randomly delete data points from the majority class.

5
New cards

SMOTE

__ operates in the feature space and creates new data points from the minority class.

6
New cards

The generator

__ tries to mimic examples from a training dataset by transforming a random source of noise received as input into a synthetic sample.

7
New cards

cost matrix

In cost-sensitive classification, the probability threshold can be changed using a __.

8
New cards

fit()

In sklearn models, the __ function has a class_weight parameter.

9
New cards

Precision

__ = TP / (TP + FP)

10
New cards

Recall

__ = TP / (TP + FN)

11
New cards

Specificity

__ = TN / (TN + FP)

12
New cards

prejudice, favoritism

Fairness is the absence of any or towards an individual or a group based on their intrinsic or acquired traits in the decision-making context.

13
New cards

Demographic

__ groups are determined based on sensitive attributes.

14
New cards

positive

The instances in both unprivileged and privileged groups should have equal probability to receive __ outcomes.

15
New cards

True Positive, False

EO states that instances from privileged and unprivileged groups should have equal Rate (TPR) and Positive Rate (FPR).

16
New cards

Pre-Processing, In-Processing, Post-Processing

Three types of Mitigation Algorithms are: __, __, and __.

17
New cards

Pre-Processing

__ techniques pre-process the dataset only by using strategies such as fairness through "unawareness", preferential sampling, massaging, and reweighing.

18
New cards

In-Processing

__ techniques adjust/tune the classification algorithm during the model training, and are dependent on the implemented classifier.

19
New cards

Post-Processing

__ techniques eliminate the discrimination from the final predictions and change the predicted outcomes of classifiers based on certain rules or constraints.