metrics.json

{
  "dataset": "aigovdev/ai-governance-scenarios",
  "examples": 12,
  "target_distribution": {
    "lower": 6,
    "high": 3,
    "unacceptable": 3
  },
  "evaluation": {
    "method": "stratified_3_fold_cross_validation",
    "accuracy": 1.0,
    "macro_precision": 1.0,
    "macro_recall": 1.0,
    "macro_f1": 1.0,
    "confusion_matrix_labels": [
      "lower",
      "high",
      "unacceptable"
    ],
    "confusion_matrix": [
      [
        6,
        0,
        0
      ],
      [
        0,
        3,
        0
      ],
      [
        0,
        0,
        3
      ]
    ]
  },
  "features": [
    "sector",
    "impact",
    "decision_autonomy",
    "human_oversight",
    "monitoring",
    "traceability",
    "technical_documentation"
  ],
  "risk_mapping": {
    "low": "lower",
    "limited": "lower",
    "high": "high",
    "unacceptable": "unacceptable"
  },
  "runtime": {
    "scikit_learn": "1.9.0"
  },
  "limitations": [
    "Dataset contains only 12 synthetic scenarios.",
    "Evaluation is illustrative and not a production benchmark.",
    "Risk tiers are engineering labels, not legal classifications."
  ]
}