Loan Approval Simulator Game

Create fictional applicant datasets, train classifiers, tune thresholds, inspect fairness, explain predictions, handle drift, and improve responsible lending simulations through interactive challenges and feedback.

Guided tutorial Level 1: Build a basic classifier
Game score0
Compute credits100
Cleaning actions12
Training attempts8
Objective: Generate data and train your first model.0%

Mission control

Choose a learning mode, difficulty, level, and fictional operating objective.

12 points
8 points
15%
Start with a balanced dataset, keep useful financial features, and compare validation performance before deployment.

Responsible-use rules

Educational simulation only.
  • Every applicant and outcome is fictional.
  • Predictions must not support real lending decisions.
  • Real systems require legal, ethical, security, and expert review.
  • Fairness cannot be guaranteed by removing one attribute.

How scoring works

Validation performance35%
Fairness and calibration25%
Error-cost control20%
Efficiency and explainability20%

Interactive application flow

Fabric.js renders the current fictional application, preparation pipeline, probability meter, and decision outcome.

Fictional dataset builder

Generate, edit, import, clean, and inspect applicant records.

180 records
8%
4%
2%

Data preparation

75% train

Feature selection and engineering

Choose model inputs. Sensitive fictional grouping stays available for fairness auditing but is excluded from training by default.

Applicant records

No dataset generated

Classifier training lab

Train an actual browser-based educational classifier using the selected fictional attributes.

0.55
±0.07
0.012
160
6
0.015
Iteration0
Training loss
Validation loss
Training accuracy
Validation accuracy
Generalisation gap

Training log

Generate and prepare a dataset before training.

Evaluation and threshold challenge

Inspect metrics, confusion patterns, calibration, feature influence, and individual decisions.

Explainable prediction

Train a model to inspect a fictional applicant.
Counterfactual suggestions will appear after evaluation.

Error analysis detective

IDActualPredictionProbabilityPattern
No evaluated errors yet.

Model comparison tournament

ModelAccuracyPrecisionRecallF1AUCFairness gapComplexityRecommendation

Fairness and bias lab

Audit fictional groups, test mitigation options, and learn why proxy variables can preserve unfair patterns.

Largest approval gap
Largest FPR gap
Largest FNR gap
Fairness score
Underrepresented group
Proxy warning

Audit notes

Train and evaluate a model before running the audit.

Responsible checks

Production and concept-drift simulation

Change the fictional applicant population, monitor decay, and choose a responsible response.

0%
0%
0%
0%
0%

Response decision

Simulate drift to receive a response recommendation.

Monitoring alerts

PeriodMetricValueStatusSuggested action

Final game report

Review performance, fairness, efficiency, explanations, and deployment readiness.

Final score0
Selected model
Best threshold
Validation F1
Fairness score
ReadinessNot assessed

Executive summary

Build the report after training, evaluation, fairness auditing, and drift simulation.

Improvement recommendations

    Decision checklist

    Session details

    Related Calculators

    Spam Email ClassifierFruit Classification GameWeather Type PredictorCustomer Churn ChallengeDisease Risk ClassifierAnimal Species ClassifierSentiment Classification GameHandwritten Digit Challenge

    Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.