Toolkit chevron_right Explainability Lab

Model Explainability Lab

Use SHAP and LIME to understand how models make predictions.

dark_mode

edit_note Input Text

LIME Explanation

Word colors show contribution: green = positive, red = negative

bar_chart SHAP Feature Importance

category Aspect-Based Sentiment

compare_arrows Counterfactual Explanations

What minimal changes would flip the prediction?

Try:

gavel Defendant Profile

Based on ProPublica's COMPAS analysis. Demonstrates how ML explainability reveals racial bias.

Age30

Prior Offenses2

Charge Severity

Gender

Race

Juvenile Felonies0

waterfall_chart SHAP: Why This Score?

compare_arrows Fairness Comparison: Same Profile, Different Race

school Why This Matters

ProPublica's 2016 investigation found that COMPAS was biased against Black defendants — they were nearly twice as likely to be falsely flagged as high risk compared to White defendants. Explainability tools like SHAP help reveal when race acts as a hidden driver of predictions, even when the model claims to be "fair."

Model Explainability Lab

edit_note Input Text

LIME Explanation

bar_chart SHAP Feature Importance

category Aspect-Based Sentiment

compare_arrows Counterfactual Explanations

tune Applicant Features

waterfall_chart SHAP Waterfall

tune Patient Features

waterfall_chart SHAP Feature Importance

gavel Defendant Profile

warning Racial Bias Detected

waterfall_chart SHAP: Why This Score?

compare_arrows Fairness Comparison: Same Profile, Different Race

school Why This Matters