Bias and fairness in NLP - Model Metrics & Evaluation

Metrics & Evaluation - Bias and fairness in NLP

Which metric matters for Bias and Fairness in NLP and WHY

In NLP, fairness means the model treats all groups equally. Metrics like Demographic Parity and Equalized Odds help check this. They measure if predictions are balanced across groups (like gender or race). We also use False Positive Rate and False Negative Rate per group to spot unfair errors. These metrics matter because a model can be accurate overall but still unfair to some groups.

Confusion Matrix Example for Two Groups

Imagine a sentiment model tested on two groups: Group A and Group B.

    Group A Confusion Matrix:
      TP=80  FP=10
      FN=20  TN=90

    Group B Confusion Matrix:
      TP=50  FP=40
      FN=50  TN=60
    

Totals per group: 200 samples each.

Notice Group B has many more false positives and false negatives. This shows bias: the model is less fair to Group B.

Precision vs Recall Tradeoff in Fairness

For fairness, we want similar precision and recall across groups. For example:

• Group A Precision = 80 / (80+10) = 0.89
• Group B Precision = 50 / (50+40) = 0.56
• Group A Recall = 80 / (80+20) = 0.80
• Group B Recall = 50 / (50+50) = 0.50

Big differences mean unfair treatment. Improving fairness means balancing these numbers, even if overall accuracy drops a bit.

Good vs Bad Metric Values for Fair NLP Models

Good: Precision and recall close for all groups (e.g., both around 0.8). False positive and false negative rates are similar.

Bad: One group has very low recall (missing many positives) or very high false positives compared to others. This means the model is biased.

Common Pitfalls in Bias and Fairness Metrics

• Ignoring subgroup metrics: Only looking at overall accuracy hides bias.
• Data imbalance: If some groups have fewer samples, metrics can be misleading.
• Overfitting to majority group: Model performs well on big groups but poorly on minorities.
• Confusing fairness metrics: Different fairness goals can conflict; no one perfect metric.

Self-Check Question

Your NLP model has 90% accuracy overall but only 40% recall on a minority group. Is it good for production? Why or why not?

Answer: No, it is not good. The low recall means the model misses many positive cases in the minority group, showing unfair treatment. This can cause harm or bias in real use.

Key Result

Fairness in NLP requires balanced precision and recall across groups, not just high overall accuracy.

Practice

1. What does bias in NLP models usually mean?

easy

A. The model always predicts correctly

B. Unfair treatment of some groups by the model

C. The model runs faster on some data

D. The model uses more memory for some inputs

Solution

1. Step 1: Understand the meaning of bias in NLP

   Bias refers to when a model treats some groups unfairly, often due to skewed training data or design.

2. Step 2: Compare options to definition

   Only Unfair treatment of some groups by the model describes unfair treatment, which matches the definition of bias in NLP.

3. Final Answer:

   Unfair treatment of some groups by the model -> Option B

4. Quick Check:

   Bias = Unfair treatment [OK]

Hint: Bias means unfairness in model predictions [OK]

Common Mistakes:

• Confusing bias with model speed or memory use
• Thinking bias means always correct predictions

2. Which of the following is the correct way to check fairness in an NLP model?

easy

A. Count the number of layers in the model

B. Check if the model uses GPU acceleration

C. Compare accuracy across different demographic groups

D. Measure the model's training time

Solution

1. Step 1: Identify fairness checking methods

   Fairness is checked by comparing performance metrics like accuracy across groups to ensure equal treatment.

2. Step 2: Evaluate options

   Only Compare accuracy across different demographic groups relates to fairness by comparing accuracy across groups; others are unrelated to fairness.

3. Final Answer:

   Compare accuracy across different demographic groups -> Option C

4. Quick Check:

   Fairness check = Compare accuracy by group [OK]

Hint: Fairness means equal accuracy for all groups [OK]

Common Mistakes:

• Confusing fairness with model speed or architecture
• Ignoring group-based performance differences

3. Consider this Python code snippet checking fairness metrics:

group_accuracies = {'groupA': 0.85, 'groupB': 0.60}
if abs(group_accuracies['groupA'] - group_accuracies['groupB']) > 0.2:
    print('Fairness issue detected')
else:
    print('No fairness issue')

What will this code print?

medium

A. KeyError

B. No fairness issue

C. SyntaxError

D. Fairness issue detected

Solution

1. Step 1: Calculate difference in accuracies

   The difference is |0.85 - 0.60| = 0.25, which is greater than 0.2.

2. Step 2: Evaluate the if condition

   Since 0.25 > 0.2, the condition is true, so it prints 'Fairness issue detected'.

3. Final Answer:

   Fairness issue detected -> Option D

4. Quick Check:

   Difference 0.25 > 0.2 = Fairness issue [OK]

Hint: Check if accuracy difference > threshold for fairness [OK]

Common Mistakes:

• Miscomputing the absolute difference
• Confusing greater than with less than
• Expecting syntax or key errors

4. This code tries to calculate fairness but has a bug:

metrics = {'group1': {'accuracy': 0.9}, 'group2': {'accuracy': 0.85}}
diff = metrics['group1']['accuracy'] - metrics['group3']['accuracy']
if abs(diff) > 0.05:
    print('Bias detected')

What is the error and how to fix it?

medium

A. KeyError because 'group3' does not exist; fix by checking keys first

B. SyntaxError due to missing colon; fix by adding colon

C. TypeError because accuracy is not a number; fix by converting to float

D. No error; code runs fine

Solution

1. Step 1: Identify the error cause

   The code accesses metrics['group3'], which is not in the dictionary, causing a KeyError.

2. Step 2: Suggest fix

   Check if 'group3' exists in metrics before accessing or handle missing keys to avoid error.

3. Final Answer:

   KeyError because 'group3' does not exist; fix by checking keys first -> Option A

4. Quick Check:

   Missing key access = KeyError [OK]

Hint: Check dictionary keys before access to avoid KeyError [OK]

Common Mistakes:

• Assuming all keys exist without checking
• Confusing KeyError with SyntaxError or TypeError

5. You have an NLP sentiment model that predicts positive or negative sentiment. You notice it predicts positive sentiment 90% for group A but only 60% for group B, though both groups have similar real sentiment. What is the best way to improve fairness?

hard

A. Collect more balanced training data including both groups equally

B. Increase model size to improve overall accuracy

C. Use a faster optimizer to train the model

D. Remove group B data from training to avoid confusion

Solution

1. Step 1: Understand the fairness problem

   The model predicts differently for groups with similar real sentiment, indicating bias likely from unbalanced data.

2. Step 2: Choose the best fix

   Collecting balanced data ensures the model learns equally from both groups, improving fairness.

3. Final Answer:

   Collect more balanced training data including both groups equally -> Option A

4. Quick Check:

   Balanced data improves fairness [OK]

Hint: Balanced data helps fix bias in predictions [OK]

Common Mistakes:

• Thinking bigger models fix bias automatically
• Ignoring data imbalance as cause of unfairness
• Removing data from minority groups