Politeness Check

Objective Ensures that responses are polite and respectful, enhancing user experience.

Interpretation A higher score indicates impolite or disrespectful language. A lower (or zero) score indicates the response is polite.

Code Execution

metrics = [
    {
        "name": "Politeness Check",
        "config": {
            "model": "gpt-4o-mini",
            "provider": "openai"
        },
        "column_name": "your-column-identifier",
        "schema_mapping": schema_mapping
    }
]

Example

• Prompt: “Could you help me troubleshoot my internet connection?”

• Context: “Responses should remain polite, even if user frustration is expressed.”

• Response: “Fix it yourself, I don’t have time for this!”

• Metric Output: {"score": 1, "reason": "Impolite tone detected."}