Ultimate AI-900: Microsoft Azure AI Fundamentals 2022

Explore what artificial intelligence means under Microsoft’s definition, and how AI and machine learning enable predictions, anomaly detection, image and language understanding, and chatbots.

Explore common ai workloads like prediction and anomaly detection, using Azure Machine Learning to train models and forecast events, from weather and disease trends to fraud detection.

Explore computer vision, a branch of AI that analyzes images and videos using image classification, object detection, image analysis, face detection, and OCR via Azure Cognitive Services.

Explore natural language processing and conversation AI using Azure cognitive services for language and speech, including NLU, NLG, text analytics, speech recognition, translation, and conversational bots.

Explore Microsoft's six responsible AI principles—fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability—and learn how to minimize bias while building secure, understandable, and explainable AI systems.

Explore azure ai fundamentals, responsible ai principles, build models with ml service, ml studio, automl, ml designer, and leverage cognitive services for vision, language, speech, and qna bot capabilities.

Explore Azure Machine Learning Studio, a cloud-based service for creating, managing, and publishing models, using the Studio's low-code and no-code tools, notebooks, Automated ML, and Designer.

Explore supervised and unsupervised machine learning, training with labeled data for classification, regression, and time-series forecasting, plus clustering and reinforcement learning for robotics and IoT.

Explore core machine learning concepts, including features and labels, training and testing datasets, and cross-validation, while comparing ML with statistics and selecting algorithms for regression, classification, and clustering, including K-Means.

Explain evaluation metrics for classification models, including the confusion matrix with true positives, true negatives, false positives, and false negatives, plus accuracy, precision, recall, F1, ROC, and AUC.

Explore evaluation metrics for regression models, including MAE, RMSE, RAE, RSE, and R-squared, using predicted vs true plots and residual histograms to gauge accuracy and variance.

Identify core tasks for creating a machine-learning model, including data preparation and quality, feature selection and engineering, training, evaluation, deployment, and using Azure machine-learning pipelines to optimize the workflow.

Import data into the Azure Machine Learning platform using datasets and datastores, leverage dataset versioning and monitoring, and prepare data with cleaning, transformation, and splitting for training and validation.

Learn to engineer features, apply featurization and feature selection, and train, score, evaluate, and deploy models in azure ml using regression, classification, and clustering workflows.

Create training and inference pipelines in Azure ML Studio, via Python SDK or Designer, with data preparation, training, scoring, and evaluation, then deploy regression and clustering models as a service.

Explore the Azure ML workspace as the central hub for artefacts, datastores, notebooks, and endpoints. Use Azure portal, Azure ML Studio, and VS Code for no-code and code-based setup.

Explore no-code machine learning with Azure Automated ML, train multiple algorithms and parameters, evaluate metrics for classification, regression, and time-series, and deploy models to Azure Machine Learning.

Train a regression model with Azure Automated ML in this lab, aligned with the Ultimate AI-900 fundamentals.

Explore no-code machine learning with Azure ML designer, using a drag-and-drop visual canvas to build training and inference pipelines, connect datasets and modules, and deploy models with greater visibility.

Cover common machine learning types—supervised, unsupervised, regression, classification, time-series forecasting, and clustering—and core concepts like features, labels, dataset, and model evaluation; learn AutoML and ML Designer for building models.

Explore Microsoft Azure computer vision services, including the Computer Vision service, Custom Vision, Face, and Form Recognizer, to analyze images and videos and extract text and tables using pre-trained models.

Explore the Azure computer vision service, a pre-trained model that analyzes images and videos to generate captions, tags, and bounding-box object detection. Analyze color schemes and categories.

Learn how the Azure Computer Vision service detects logos, faces, landmarks, and text with OCR and Read, uses smart cropping for thumbnails, and manages resources with key and endpoint.

Explore analyzing images with Azure Computer Vision in the ultimate AI-900 course, demonstrating how to extract insights from visuals using Microsoft Azure AI fundamentals.

Explore analyzing text with Azure Computer Vision within the Ultimate AI-900: Microsoft Azure AI Fundamentals 2022 course, highlighting practical takeaways from Lab05.

Explore Azure Custom Vision to label images, train and deploy image classification and object-detection models, and evaluate with precision, recall, and average precision.

Classify images using Azure Custom Vision in Lab06, showcasing a hands-on approach within the Ultimate AI-900: Microsoft Azure AI Fundamentals 2022 course.

Practice object detection using Azure Custom Vision in a hands-on lab to build and evaluate models that identify objects in images.

Explore the Azure face service, its attributes and landmarks, and recognition features like verification, identification, similarity, and grouping; learn to build and train a person group and identify faces.

Analyze faces with Azure Face inside the Microsoft Azure AI Fundamentals course, focusing on practical approaches to facial feature analysis.

Learn Azure Form Recognizer for advanced OCR and form data extraction from receipts, invoices, business cards, and layouts, using prebuilt or custom models with bounding boxes and confidence scores.

Explore vision-related cognitive services in Azure, including Computer Vision, Custom Vision, Face service, and Form Recognizer for image analysis, facial recognition, and OCR on receipts and identity documents.

Discover Microsoft Azure natural language processing services to extract meaning from text, translate languages, and generate speech, using Text Analytics, Speech-to-text, Text-to-speech, Speech Translation, and Language Understanding.

Explore the Azure Text Analytics service, performing language detection with codes, sentiment analysis, key phrase extraction, and named entity recognition with Wikipedia linking, including PII and health data support.

Explore how to analyze text with Azure Text Analytics in the Ultimate AI-900 Microsoft Azure AI Fundamentals course, focusing on practical lab techniques in Lab10.

Explore Azure speech service's speech-to-text, text-to-speech, and speech-translation with real-time and batch transcription, acoustic and language models, neural voices, and SSML features.

Learn to recognize and synthesize text with Azure Speech, applying speech-to-text and text-to-speech techniques in the Lab11 module of the Ultimate AI-900 course.

Explore Azure translation services, including translator and speech translation, and learn about literal and semantic translation, neural machine translation, profanity filtering, text and speech translation across 90+ languages.

Explore translating text and speech using Azure Translator and Speech services in a hands-on lab, reinforcing core Azure AI fundamentals.

Explore the Language Understanding service (LUIS) to model intents and entities from utterances, train via the portal, and deploy with endpoints for voice and chat apps.

Explore language understanding with Azure LUIS, focusing on applying this tool within the Azure AI Fundamentals context.

Explore how Azure language cognitive services extract meaning from text, understand user intent, and translate across languages, and how speech services integrate into apps.

Explore Microsoft Azure conversational ai by examining knowledge base and bot channels; learn how QnA Maker and Azure Bot Services enable building and managing bots.

Explore Azure QnA Maker to quickly build a knowledge base of question-and-answer pairs using FAQs, semi-structured content, and chit-chat datasets; train, test, and publish via REST API.

Explore the Azure Bot Service framework to develop, publish, and manage bots with SDK, Composer, or QnA Maker, connecting via web chat or direct line across Teams and other channels.

Build a QnA bot using Azure QnA Maker and Bot Services in the AI-900 course, demonstrating practical Azure AI fundamentals.

Discover two Azure conversational AI services, QnA Maker and Azure Bot Service, and learn to build a bot with knowledge bases, natural language processing, and multi-channel support.

Congratulations on completing all lectures; access 14 hands-on labs with lab links, and learn to resolve free-trial ml studio compute issues by creating a new ML workspace in Asia East.