Azure AI Language: NLP Guide

Key Features and Learning Objectives

• Language Detection: Identifies languages (e.g., English, Spanish).
• Key Phrase Extraction: Pulls important terms (e.g., “New York”).
• Sentiment Analysis: Scores text as positive, negative, or mixed.
• Entity Extraction: Finds people, places, organizations.
• Text Summarization: Condenses long documents.
• Translation: Supports 100+ languages.

Sentiment Analysis Example

Analyzing Text and Sentiment

pythonCopy
from azure.ai.textanalytics import TextAnalyticsClient
client = TextAnalyticsClient(endpoint, credential)
result = client.analyze_sentiment(documents=["Great service!"])

Entity Extraction and Linking

jsonCopy
{
  "documents": [
    {
      "text": "Visit New York",
      "entities": [
        {
          "text": "New York",
          "category": "Location",
          "url": "https://en.wikipedia.org/wiki/Manhattan"
        }
      ]
    }
  ]
}

Text Summarization

jsonCopy
{
  "kind": "ExtractiveSummarization",
  "documents": [
    {
      "text": "First sentence summarizing key points..."
    }
  ]
}

Building Conversational AI

Translation and Speech

Connecting Apps to Azure AI Language

csharpCopy
var client = new TextAnalyticsClient(endpoint, credential);
var response = client.DetectLanguage(new List<string> { "Hello, world!" });