Implement text analysis solutions — Microsoft Certified: Azure AI Apps and Agents Developer Associate

Official references for this objective

• Microsoft — Evaluation metrics | Microsoft Learn

  Partial Ratio: Calculates the similarity by taking the shortest string, and compares it against the sub-strings of the same length in the longer string.

• Microsoft — Evaluating the performance of LLM summarization prompts with G-Eval | Microsoft Learn

  GPT-4 does not support the output of token probabilities. G-Eval sets n=20 , temperature=2 , top_p=1 to sample 20 times to estimate the token probabilities.