AI-900: How to interpret unknown language detection results in Text Analytics service

Learn what it means when the Text Analytics service returns unknown as the language name and identifier, and NaN as the score, and how to handle ambiguous or mixed language texts.

Question

After analyzing a text with Text Analytics service, the service returned that the language detected is unknown, with a score of NaN. What does this mean?

Select all options that apply.

A. Text is ambiguous in nature
B. Analysis of text failed
C. Text has mixed language content

Answer

A. Text is ambiguous in nature
C. Text has mixed language content

Explanation

When results return a value of unknown for the language name and the language identifier, and a score of NaN, it means that text is ambiguous in nature or has mixed language content.

Answer: A and C are correct.

Explanation: The Text Analytics service can detect the language of a given text by using a pre-trained model that assigns a score to each possible language. The score reflects the confidence of the model that the text belongs to that language. The service returns the language name, the language identifier, and the score for each document in the input. However, sometimes the service may not be able to detect the language of a text with high confidence, and it will return a value of unknown for the language name and the language identifier, and a score of NaN (not a number). This can happen for two reasons:

• The text is ambiguous in nature, meaning that it does not contain enough distinctive words or features that can help the model identify the language. For example, a text that consists of only numbers, symbols, or common words that are shared by many languages may be ambiguous for the model.
• The text has mixed language content, meaning that it contains words or phrases from more than one language. For example, a text that switches between English and French may have mixed language content. The model may not be able to assign a single language to the whole text, and it will return unknown as the result.

Therefore, the correct options are A and C. Option B is incorrect because the analysis of the text did not fail, it just did not produce a definitive result.

Microsoft Azure AI Fundamentals AI-900 certification exam practice question and answer (Q&A) dump with detail explanation and reference available free, helpful to pass the Microsoft Azure AI Fundamentals AI-900 exam and earn Microsoft Azure AI Fundamentals AI-900 certification.