Azure AI Language text analysis: language detection, key phrases, entities, entity linking and sentiment with opinion mining

Azure AI Language offers prebuilt natural language processing (NLP) features that analyze text without any training. You send documents (plain text with an id and optionally a language code) and choose the kind of analysis. The REST call posts to the language/:analyze-text path with a kind such as LanguageDetection, KeyPhraseExtraction, EntityRecognition, EntityLinking or SentimentAnalysis; the SDKs expose methods such as detect_language and analyze_sentiment. Results come back per document, with errors reported per document so one bad input does not fail the batch. Larger jobs, and combining several analyses in one request, use the asynchronous analyze-text jobs endpoint.