FREE Microsoft Azure AI Fundamentals (AI-900) Questions and Answers
Which of the following does not fall within the category of LUIS application intents?
In a LUIS (Language Understanding) application, "Filter" is not an entity type associated with intents. In LUIS, intents represent the actions or tasks that users express in their input. Entities, on the other hand, are specific pieces of information within the user's input that provide context to the intent. Common entity types in LUIS include things like dates, times, locations, quantities, and more, which help extract specific details from user input. However, "Filter" is not a standard entity type in LUIS's built-in entity recognition system.
What are the two specialized domain models of computer vision?
Please select 2 correct answers
These specialized domain models are designed to enhance the capabilities of the Computer Vision service for specific recognition tasks related to celebrities and landmarks. However, keep in mind that Azure's services and offerings may have evolved since my last update, so I recommend checking the latest Azure documentation or portal for the most current information.
Which of the following resources cannot be produced using Azure Machine Learning Studio?
Azure Machine Learning Studio allows you to create and manage various resources related to machine learning, such as datasets, experiments, pipelines, trained models, and compute resources. However, "Compute Balancers" is not a standard resource type available in Azure Machine Learning Studio. Instead, you can create compute resources like compute clusters, virtual machines, and other compute instances to run your machine learning experiments and processes.
What method is used to group comparable entities based on features?
Clustering is an unsupervised machine learning technique that involves grouping data points or entities into clusters based on their similarity. The goal is to identify patterns in the data and group together entities that share common characteristics. Clustering algorithms analyze the relationships between data points and create clusters where the entities within each cluster are more similar to each other than to those in other clusters. It's commonly used for tasks like customer segmentation, image segmentation, and data exploration.
Which of these metrics is not applied to Custom Vision model training?
The F1 Score is a metric that combines Precision and Recall into a single value and is commonly used in binary classification tasks. While it's a valuable metric to consider, it might not be directly used in the Custom Vision interface for model training evaluation.
What resource kinds are offered as part of the computer vision service?
"Computer Vision" is a field of artificial intelligence that involves training computers to interpret and understand visual information from the world, such as images and videos. "Computer Services" is a broad term that can refer to various IT services provided by computers or the cloud.
Which image format in the Face Service is not supported?
The term "AI" is not an image format. It stands for "Artificial Intelligence." As for supported image formats in the Azure Face service, common image formats such as JPEG, PNG, GIF, and BMP are supported. These formats are used for providing input images to the service for facial recognition and related tasks. If you are referring to a specific image format abbreviation "AI," it's not a recognized image format in the context of the Azure Face service.
Which one is a multiclass classification algorithm?
A Decision Forest is a type of ensemble learning algorithm that consists of multiple decision trees. Each decision tree is trained on a subset of the data and makes its own predictions. In multiclass classification tasks, a decision forest combines the outputs of multiple decision trees to determine the class label for an input data point. Each tree's prediction contributes to the final decision by a voting or averaging mechanism, allowing the algorithm to handle multiple classes effectively.
With the aid of this computer vision approach, text may be found and read within images.
OCR is a technology that involves the conversion of different types of documents, such as scanned paper documents, PDF files, or images taken by a digital camera, into editable and searchable data. It extracts text from images, making it possible to analyze, edit, and store the text as digital content. OCR is widely used in various applications, including digitizing printed documents, extracting data from images for further analysis, converting scanned documents into machine-readable formats, and enabling accessibility features for visually impaired individuals by converting text into audio output.
All of the information listed here, with the exception of:
The OCR (Optical Character Recognition) API can return hierarchical information about text extracted from images. However, it does not typically return information about "Areas." The OCR API is primarily used to extract text and recognize characters from images, providing details such as recognized text, bounding boxes for text regions, and the layout of the text in the image. It can give information about text lines, words, and characters, but the concept of "Areas" is not a standard output from the OCR API.
Which of these can be utilized to develop conversational AI?
Q&A Maker is a cognitive service provided by Microsoft Azure that allows you to easily build, train, and deploy a question-and-answer-based chatbot. It takes a set of questions and corresponding answers and uses them to create a chatbot that can respond to user queries in a conversational manner. By using Q&A Maker, you can build chatbots that provide instant answers to frequently asked questions and engage users in natural language conversations. It's a valuable tool for creating customer support bots, virtual assistants, and other conversational AI applications.
Which would be utilized to determine the category to which an item belongs?
Classification is a supervised machine learning task that involves assigning a label or category to an input data point based on its features. In the context of predicting categories for items, classification algorithms analyze the characteristics of the items and learn to classify them into predefined categories or classes. This is a common task in various fields, such as image recognition, text categorization, and customer segmentation, where you want to determine which category an item best fits into based on its attributes.
In this computer vision challenge, each pixel in an image is categorized according to the object to which it belongs.
Semantic segmentation is a technique in computer vision where each pixel in an image is assigned a label corresponding to the object or class it belongs to. It's used to achieve a detailed understanding of the image's content by segmenting it into different regions based on the semantics of the objects present.
This is done to draw attention to the strongest trend in a dataset:
PCA is a dimensionality reduction technique commonly used in data analysis and machine learning. Its primary purpose is to transform a high-dimensional dataset into a lower-dimensional representation while retaining as much of the original data's variance as possible. The first principal component captures the direction in which the data varies the most, and subsequent components capture orthogonal directions of decreasing variance.
Which classification doesn't fall within machine learning algorithms?
Normalization is a data preprocessing technique used to scale features within a dataset to a standard range, typically between 0 and 1. It's not a distinct machine learning algorithm category but rather a step in preparing data for machine learning algorithms.