AI-900-7

 

Module : 7 Get started with natural language processing in Azure


Introduction to Natural Language Processing (NLP)

Natural Language Processing (NLP) is a part of Artificial Intelligence that allows machines to understand, interpret, and respond to human language. Its main goal is to draw meaning and structure from text.




Real-World Applications of NLP

Customer Feedback Analysis:  
NLP helps businesses examine large amounts of customer reviews, support tickets, and surveys. By using sentiment analysis and key phrase extraction, companies can spot dissatisfaction early and improve customer experience.

Healthcare Text Analysis:  
In healthcare, Azure’s language tools pull clinical details from unstructured medical documents. Features like entity recognition and text analytics for health help identify symptoms, medications, and diagnoses. This enables faster and more accurate decisions.

Conversational AI with Virtual Agents:  
Azure language solutions also support virtual assistants that understand user intent, translate conversations, extract entities, and respond intelligently.




Understanding Natural Language Processing (NLP) on Azure

Natural Language Processing (NLP) allows machines to understand and analyze human language. Key NLP tasks include:

Language Detection

Sentiment Analysis

Named Entity Recognition

Text Classification

Translation

Summarization

Azure AI Services for NLP

Azure AI Language Service: This is a cloud-based tool that helps analyze and understand text. It supports sentiment analysis, key phrase extraction, text summarization, and conversational language understanding.

Azure AI Translator Service: This is a cloud-based translation service that uses Neural Machine Translation (NMT). It understands the meaning of text to provide more accurate and natural translations.



Understanding Azure AI Language’s Text Analysis Capabilities

Azure AI Language is a powerful cloud-based service that is part of Microsoft’s Azure AI offerings. It provides strong Natural Language Processing (NLP) capabilities to help organizations analyze, understand, and gain insights from unstructured text data like reviews, documents, emails, or social media posts.

The service includes several key text analysis features, such as entity recognition, sentiment analysis, key phrase extraction, language detection, summarization, and more. These features are essential for businesses that want to turn text into useful information.

Core Features of Azure AI Language

1. Named Entity Recognition (NER)

The Named Entity Recognition feature identifies specific entities within text, such as people, places, organizations, quantities, or dates. Azure AI Language can also be customized to extract entities from specific categories based on the industry or business needs.

Entities can be categorized by type and, in some cases, by subtype. Below are a few examples:



Type             SubType         Example  
Person                 "Bill Gates", "John"  
Location                 "Paris", "New York"  
Organization         "Microsoft"  
Quantity     Number           "6" or "six"  
Quantity     Percentage "25%" or "fifty percent"  
Quantity     Ordinal         "1st" or "first"  
Quantity     Age         "30 years old"  
Quantity     Currency "$10.99"  
Quantity     Dimension "10 miles", "40 cm"  
Quantity     Temperature "45 degrees"  
DateTime     Date          "May 2nd, 2017"  
DateTime     Time         "8am"  
DateTime     DateRange "May 2nd to May 5th"  
DateTime     Duration "1 minute and 45 seconds"  
URL                         "https://www.bing.com"  
Email                 "support@microsoft.com"  
US-based Phone Number "(312) 555-0176"  
IP Address                   "10.0.1.125"  


2. Entity Linking

The Entity Linking feature connects recognized entities to a specific reference source, like Wikipedia. This helps clarify entities, such as distinguishing between “Paris, France” and “Paris, Texas.”

Example:  
Text: “I ate at the restaurant in Seattle last week.”



3. Personal Identifying Information (PII) Detection

Azure AI Language includes PII detection, which identifies sensitive or personally identifiable information, including Personal Health Information (PHI). This is crucial for industries like healthcare or finance, where privacy is important. PII detection can find items such as:

- Names and addresses
- Email IDs and phone numbers
- Credit card details
- IP addresses
- Health-related information

4. Language Detection

The Language Detection feature identifies the language of the text and provides both the language name and its ISO 6391 code (for example, “en” for English or “es” for Spanish). It also includes a confidence score to show how certain the model is about its prediction.

Example:  
If you receive customer feedback in multiple languages, Azure AI Language can automatically identify them:

Review 1: "A fantastic place for lunch. The soup was delicious."

Review 2: "Comida maravillosa y gran servicio."

Review 3: "The croque monsieur avec frites was terrific. Bon appetit!"



Even though Review 3 contains both English and French, Azure detects English as the main language based on phrase length and content ratio.

In cases where the text is unclear or contains minimal input (for example, “:-)”), the service might return “unknown” for the language name and “NaN” (Not a Number) for the score.

5. Sentiment Analysis and Opinion Mining

Sentiment Analysis determines the emotional tone of the text, whether it’s positive, neutral, or negative. This feature is especially useful for analyzing social media comments, customer reviews, or feedback from support.

Azure AI Language uses a prebuilt machine learning model that assigns a score between 0 and 1 for each sentiment category.

Example 1 (Positive Review):

“We had dinner at this restaurant last night and the staff was courteous. The food was amazing.”

Results:

Document Sentiment: Positive  

Positive Score: 0.90  
Neutral Score: 0.10  
Negative Score: 0.00  

Example 2 (Negative Review):

“The service was slow, and the food was awful. I’ll never eat here again.”

Results:

Document Sentiment: Negative  

Positive Score: 0.00  
Neutral Score: 0.00  
Negative Score: 0.99  

This helps businesses quickly identify satisfied or unhappy customers and take action as needed.

6. Key Phrase Extraction

Key Phrase Extraction identifies the main ideas or themes from unstructured text. This helps summarize large amounts of data and discover common topics or trends.

Example:  
Review:

“We had dinner here for a birthday celebration and had a fantastic experience. The food was amazing, and the service was terrific.”

Extracted Key Phrases:

- birthday celebration
- fantastic experience
- great food
- attentive service
- dinner
- ambiance  

This feature helps businesses efficiently analyze customer feedback without having to read every single review.

7. Summarization

The Summarization feature condenses lengthy text by identifying the most important information. It’s especially helpful for summarizing articles, reports, or support tickets, allowing users to understand key insights quickly.

Azure AI Language enables organizations to:

- Extract structured insights from unstructured data.
- Detect customer sentiment and improve satisfaction.
- Protect sensitive information through PII detection.
- Understand multilingual feedback automatically.
- Improve efficiency through summarization and automation.

From analyzing restaurant reviews to processing healthcare records, Azure AI Language helps businesses make data-driven decisions quickly and accurately.



Azure AI Language’s Conversational AI Capabilities

Azure AI Language enables intelligent, human-like conversations between AI and users through its Conversational AI features. These capabilities help businesses build chatbots, assistants, and automated systems that understand natural language, interpret meaning, and respond contextually.

What Is Conversational AI?

Conversational AI allows dialogues between humans and AI using natural language. It uses Natural Language Processing (NLP) and Machine Learning to understand context, intent, and tone, making interactions smooth and natural.

Azure AI Language includes two main conversational AI features:

Question Answering

Conversational Language Understanding (CLU)

Question Answering

The Question Answering feature lets you build conversational AI solutions that can answer natural language queries instantly and accurately. It’s ideal for chatbots or virtual assistants that respond to customer questions.

It supports multi-turn conversations and context-aware responses.

It can be integrated into websites, apps, or social media platforms.

It uses a custom knowledge base of question-and-answer pairs.

Example: A user asks a customer service bot, “What’s your return policy?” The bot finds the best-matched answer from the knowledge base and responds instantly, improving support speed and accuracy.



Conversational Language Understanding (CLU)

CLU helps AI understand the intent and meaning of phrases in a conversational setting. It can extract important entities, like names, dates, or objects, and predict what action to take.

Example: When a user says, “Turn the light off,” CLU identifies the intent (“turn off”) and the entity (“light”), then performs the action.



CLU supports command-based, enterprise, and end-to-end conversational applications. It enables:

Intent prediction (what the user wants).

Entity extraction (key details in the message).

Contextual understanding for more natural responses.

Conversational AI and Generative AI

Modern AI systems combine multiple capabilities. Conversational AI focuses on understanding and responding to input, while Generative AI extends this by creating new content. Both rely on NLP to understand human language.

Benefits of Azure AI Language Conversational Features

It understands natural, multi-turn conversations.

It delivers accurate, instant responses.

It is customizable for business-specific needs.

It easily integrates with chat and voice platforms.

It is scalable for enterprise-level interactions.



Azure AI Translator – Key Notes

Old translation issue: Early systems relied on literal translation, which led to wrong meanings, missing words, and a lack of context.

Modern AI translation: It understands semantic context, grammar, tone, and everyday language for accurate results.

Languages supported: It translates over 130 languages.

Multiple outputs: It can translate from one source language into many target languages at the same time.

Main Capabilities

Text Translation – Quick and accurate real-time text translation across all supported languages.

Document Translation – It translates multiple documents while keeping structure and formatting intact.

Custom Translation – This feature allows businesses and developers to create their own Neural Machine Translation (NMT) models.

Usage Platforms

Azure AI Foundry – This is for enterprise AI operations, model building, and app development.

Microsoft Translator Pro – A mobile app for real-time speech-to-speech translation.



Get Started in Azure AI Foundry – Key Notes

Purpose:  
Azure AI Language and Translator help add language capabilities to apps.

Ways to create solutions:

- Azure AI Foundry portal
- SDK (Software Development Kit)
- REST API

Resource Types (choose based on use case):

- Language Resource: Use only Azure AI Language services or manage billing separately.
- Translator Resource: Manage access and billing for the Translator service individually.
- Azure AI Services Resource: Use multiple Azure AI services together under one resource, allowing shared access and billing.

Ways to Create Resources:

- User interface: via Azure portal or Azure AI Foundry portal.
- Script-based creation: automated setup through code or CLI.

Choose Azure AI Foundry portal if you want to see examples and test services directly.

Azure AI Foundry Portal Overview:




A unified platform for enterprise AI operations, model building, and app development.  

Organized into Hubs and Projects:

Projects act as containers for datasets, models, and resources. This simplifies management and collaboration on AI solutions.

Playground Features:

- Language Playground: Try Azure AI Language features like sentiment analysis.
- Translator Playground: Try text translation in real-time.











Comments

Post a Comment

Popular posts from this blog

AI-900-3,4

Image
Module : 3 Get Started with Machine Learning in Azure Machine learning drives today’s AI by turning data into insights for predictions and recommendations. Creating an ML solution requires key decisions that affect cost, speed, and quality. Using Microsoft Azure, the process follows six main steps: Define the Problem: Decide what to predict and how to measure success. Get the Data: Find and access reliable data sources. Prepare the Data: Clean, explore, and format the data for modeling. Train the Model: Choose algorithms and adjust parameters. Integrate the Model: Deploy it to generate predictions. Monitor the Model: Track performance and retrain as needed. Machine learning is an ongoing process. Models evolve as data changes. Step 1: Define the Problem The first step in building a machine learning (ML) solution is to define the problem. You need to clearly understand what you want the model to predict and how you will measure success. Key Considerations Model output: What result shoul...
Read more

AI-900 12,13

Module : 12   Overview of Information Extraction * Information extraction is a process that uses AI to pull structured data from unstructured or semi-structured documents, like receipts, invoices, forms, and scanned files. It combines computer vision, OCR, and machine learning or generative AI. * Key Steps in an Information Extraction Pipeline 1. Text Detection & Extraction (OCR) - It uses computer vision to find text areas in an image. - The system extracts raw text from scanned or image-based documents. 2. Value Identification & Mapping It maps the OCR text to specific data fields using: - Machine learning - Rule-based mapping - Generative AI (in newer solutions) Example From a scanned receipt, the system extracts fields such as: - Vendor - Date - Subtotal - Tax - Total * Choosing the Right Approach When designing an information extraction solution, consider: Document Characteristics - Layout consistency - Fixed templates lead to simple, rule-based extraction - Many forma...
Read more

AI-900 10,11

Image
Module : 10  Introduction to Computer Vision Concepts  Computer Vision Tasks & Techniques   * What is Computer Vision? Computer vision uses AI techniques to process and understand visual input such as: - Images   - Videos   - Live camera streams   - It enables machines to "see" and interpret visual information.   1. Image Classification   - This predicts a single label for an entire image.   - The model is trained using many labeled images.   - For example, a smart grocery checkout can identify an apple, orange, or banana from a single item placed on a scale.   - Use case: Recognizing the main subject in an image.   2. Object Detection   - This identifies multiple objects in a single image.   It returns:   - Object labels   - Bounding box coordinates   - For example, a checkout system can detect all fruits placed together....
Read more