Master AI-102 Exam: Real Questions & Answers
Table of Contents
- Introduction
- Building a Chatbot
- Requirements
- Integrating the Chatbot
- Reducing Employee Effort in Expense Reports
- Receipt Extraction
- Azure Service for Minimizing Development Effort
- Ensuring Equitable Results in the Sales System
- Responsible AI Principles
- Monitoring the Sales System
- Containerized Anomaly Detector Deployments
- Requirements for Containerized Deployments
- Ensuring Secure and Controlled Deployments
- Training a Custom Form Recognizer Model
- Model Training Process
- Sample Files for Training
- Server-Side Encryption for Azure Cognitive Search
- Benefits of Server-Side Encryption
- Implications of Enabling Customer Managed Keys
- Predictive Maintenance with Anomaly Detector
- Monitoring Industrial Machines
- Choosing the Right Azure Service
- Streaming Speech-to-Text Solution
- Converting Speech to Text
- Method for Streaming MP3 Data
- Internet-Based Training Solution
- Language Support for Remote Learners
- Cognitive Service for Personalized Training
- Provisioning a Q&A Maker Service
- Resource Creation in a New Resource Group
- Azure Resources Created for Q&A Maker
- Building a Language Model with Language Understanding
- Adding Contributors to the Language Model
- Managing Access to Authoring Resources
- Throttling Reduction in Azure Cognitive Search
- Effective Solutions for Throttling Issues
- Migration to a High-Tier Cognitive Search Service
- Developing an Automated Call Handling System
- Recognizing Different Languages
- Verifying Learner Engagement
- Image Training for Custom Form Recognizer Model
- Training Data Set Criteria
- Selection of Files for Model Training
- Configuring Connectivity for Azure Cognitive Search
- Connecting Web App to Azure Virtual Network
- Ensuring Private Endpoint and Connectivity
- Language Model Development with Language Understanding
- Developing a Conversational Language Model
- Enabling Inclusiveness and Safety in AI Systems
- Cognitive Services Selection for Support Tickets
- Language Understanding for Incoming Emails
- Language Detection and Response in Multiple Languages
- Monitoring Factory Components with Cognitive Services
- Object Detection and Labeling for Safety Compliance
- Choosing the Right Azure Cognitive Service
- Smart Cropping Feature for Custom Vision Model
- Custom Vision Resource for Image Resizing
- Configuration of API URL for Smart Cropping
- Integrating Cognitive Insights and Video Player Widgets
- Keyword Search and Display of People
- Caption Language and Widget Integration
- Object Detection Model Deployment between Resources
- Moving Object Detection Model to Production Environment
- API Endpoints for Model Deployment
- Training Custom Vision Object Detection Model
- Actions to Perform for Model Training
- Sequence of Actions for Custom Model Deployment
- Optical Character Recognition in Images
- Performing OCR with Computer Vision
- Deployment Options for Sensitive Documents
- Adding Images to Custom Vision Classifier
- Process for Adding New Images to Classifier
- Extending and Retraining the Classifier
- Facial Recognition and Detection in Images
- Face API for Facial Recognition
- Detection of Inappropriate Images
- Performance Metrics for Custom Vision Model
- Evaluation Metrics for Model Performance
- Importance of Recall and Precision
- Performing OCR with Computer Vision Client Library
- Preventing Premature Execution of Read Operation
- Controlling Read Result Status and Operation
- Searching for Photos by Image Example
- Finding Similar Faces in Existing List
- Matching Faces and Person Verification
- Adding Multiple Images to Person Group
- Procedure for Adding Images to Person Group
- Creating Persons for Face Recognition
- Preventing Premature Execution of Read Operation
- Controlling Read Result Status
- Wrapping Get Read Result in a Loop for Delay
- Building OCR Solution for Scanning Sensitive Documents
- Options for an On-Premises Solution
- Hosting the Computer Vision Endpoint in a Container
- Updating Custom Vision Classifier with New Images
- Adding New Images to Existing Model
- Retraining the Model for Improved Accuracy
- Training Custom Vision Model with Image Data
- Efficient Training with Locally Grouped Images
- Manual Tagging and Review of Image Suggestions
- Mobile App Development with Custom Vision
- Retraining Model with Additional Images
- Extending the Model for Improved Accuracy
- Applying Optical Character Recognition in Images
- Extracting Text from Images Using Computer Vision
- Containerizing the Computer Vision Endpoint on an On-Premises Server
Building a Chatbot
Chatbots have become increasingly popular for providing automated customer support and interaction. In this section, we will explore the requirements for building a chatbot and how to integrate various services to meet those requirements.
Requirements
When building a chatbot, it is important to consider the following requirements:
- Support for chit chat and a knowledge base
- Multilingual capabilities
- Sentiment analysis on user messages
- Automatic language model selection
Integrating the Chatbot
To meet these requirements, it is recommended to integrate the following services into the chatbot:
- Language Understanding: This service allows users to Interact with the chatbot using natural language.
- Q&A Maker: A cloud-based natural language processing service that helps find the most appropriate answers from a knowledge base.
- Text Analytics: This service can be used to mine insights from unstructured text, including sentiment analysis.
- Translator: Translation service that enables multilingual capabilities.
By integrating Language Understanding, Q&A Maker, Text Analytics, and Translator into the chatbot, You can Create a comprehensive and intelligent conversational agent that can support chit chat, provide knowledge-based responses, perform sentiment analysis, and automatically select the best language model for communication.
These integrations will ensure that the chatbot meets the specified requirements and provides an enhanced user experience.