Chatbot Quality Monitoring

from ragas import evaluate
from ragas.metrics import (
    faithfulness,
    answer_relevancy,
    context_precision,
    context_recall,
)
from datasets import Dataset

# Prepare your chatbot conversation data
data = {
    "question": ["What is your return policy?"],
    "answer": ["You can return items within 30 days..."],
    "contexts": [["Our return policy allows..."]],
    "ground_truth": ["30-day return policy for unused items"]
}

dataset = Dataset.from_dict(data)

# Run evaluation
result = evaluate(
    dataset,
    metrics=[
        faithfulness,        # Is response grounded in context?
        answer_relevancy,    # Does it answer the question?
        context_precision,   # Is retrieved context relevant?
        context_recall,      # Is all needed context retrieved?
    ],
)

print(result)