API Reference

Complete reference for integrating EthixAI's fairness analysis and explainability APIs into your systems.

Core Analysis Endpoints
Primary endpoints for bias detection and fairness analysis
POST/api/analyze

Trigger a comprehensive fairness analysis on your dataset.

{
  "dataset_name": "loan_applications",
  "data": {
    "age": [25, 45, 35, ...],
    "income": [50000, 80000, 65000, ...],
    "loan_approved": [1, 1, 0, ...]
  },
  "target_column": "loan_approved",
  "protected_features": ["age", "gender"]
}
GET/api/analyses/latest

Retrieve the most recent analysis results.

GET/api/report/:id

Fetch a specific analysis report by ID.

Explainability Endpoints
Generate SHAP plots and feature importance analysis
POST/api/explain

Generate SHAP explanations for model predictions.

Compliance Endpoints
Validate against regulatory frameworks
POST/api/compliance/validate

Validate model against GDPR, ECOA, SR 11-7 frameworks.

Data Management
Upload datasets and manage analysis history
POST/api/datasets/upload

Upload a new dataset for analysis.

GET/api/datasets

List all uploaded datasets.

Authentication

All API requests require authentication via Firebase JWT tokens. Include the token in the Authorization header:

Authorization: Bearer <your-jwt-token>

Rate Limits

  • • Standard tier: 100 requests/minute
  • • Enterprise tier: 1000 requests/minute
  • • Dataset size limit: 50MB per upload