Microsoft Azure AI Fundamentals Practice Test (AI-900)

Generative models are designed to create new content, such as text or images, based on patterns learned from existing data. In this scenario, a generative model can generate unique ad copy that aligns with client needs by learning from a corpus of advertising texts.

**Classification models **categorize data into predefined classes and are not used for content creation.

Predictive analytics models forecast future data based on historical trends but do not generate new content.

Reinforcement learning models focus on decision-making processes to achieve specific goals and are not primarily used for generating creative content.

They can generate new data instances similar to the training data - This is the correct answer. Generative AI models are designed to create new data instances that resemble the training data. This makes them ideal for generating synthetic data to augment testing datasets, ensuring variety and realism in the data used for testing.

They can classify data into specific categories with high precision - This feature is characteristic of classification models, which categorize data into predefined labels but do not generate new data instances.

They can reduce data dimensionality while retaining key features - This describes dimensionality reduction techniques (such as PCA), which focus on simplifying the data without losing important features. It is not related to generating new data instances.

They can identify anomalies by learning normal data patterns - This is a feature of anomaly detection models, which are used to identify outliers or unusual patterns in data, not for generating synthetic data.

Use training data that includes a wide range of demographic groups - This is the correct answer. Ensuring that the training data is diverse and includes a wide range of demographic groups helps reduce biases and ensures that the automated loan approval system treats all applicants fairly. This approach ensures that the model learns from a broad spectrum of data, which can help minimize unfairness in the system.

Exclude any features related to personal characteristics from the data - While this may seem like a good way to prevent bias, excluding personal characteristics (such as age, gender, or race) entirely could lead to a model that lacks important context for making fair and informed decisions.

Increase the complexity of the algorithm to improve accuracy - Increasing algorithm complexity might improve accuracy, but it does not directly address fairness.

Expand the dataset by collecting more data of the same type - Expanding the dataset with more of the same type of data might not necessarily improve fairness if the data itself is biased or unrepresentative.

Identifying the individual person by matching against a database of known faces - This is the correct answer. A facial detection and analysis solution is typically focused on detecting and analyzing facial features like emotion, age, gender, head pose, and landmarks. However, identifying or recognizing a specific individual for example facial recognition by matching against a database of known faces is a different task, which requires a facial recognition solution, not just facial detection and analysis.

Detecting the presence and location of faces in an image - This is a core feature of facial detection, which is part of facial detection and analysis solutions. It allows the system to locate faces within an image.

Analyzing facial attributes such as age, gender and emotion - This is another key feature of facial analysis, which examines various facial characteristics like age, gender and emotion.

Determining facial landmarks like the position of eyes, nose and mouth - This is also part of facial analysis, as it involves detecting specific facial landmarks to understand the positioning of key facial features.

Automated Machine Learning simplifies the model development process by systematically testing multiple algorithms and hyperparameter settings to identify the best-performing model for a given dataset. It automates the time-consuming process of model selection and tuning, allowing users to focus on other tasks.

Azure Machine Learning Designer provides a visual interface to build machine learning pipelines but does not automate the selection and tuning of models.

Azure Notebooks is a cloud-based Jupyter Notebook service for writing and running code, without built-in capabilities for automated model selection.

Azure Machine Learning Interpretability offers tools to explain and interpret machine learning models but does not assist in finding the optimal model through systematic testing.

The ability to recognize and extract printed and handwritten text from images is a key feature of optical character recognition (OCR) solutions. OCR allows for the conversion of text within images into editable and searchable data formats.

The other options describe features of different computer vision solutions, classifying images into categories is related to image classification, detecting faces and analyzing features pertains to facial detection and analysis, and segmenting and identifying objects within images is a feature of object detection solutions.

The Azure AI Face Detection service is designed specifically for detecting human faces and analyzing facial features like age, emotion and gender. It provides advanced facial recognition capabilities tailored for these tasks.

The Azure AI Vision service offers general image analysis, it does not specialize in detailed facial attribute analysis.

Azure AI Form Recognizer is used for extracting text from forms and documents.

Azure AI Speech service processes and recognizes spoken language, neither of which address facial detection or analysis.

The Azure AI Language service provides pre-built natural language processing (NLP) capabilities such as key phrase extraction, entity recognition, and sentiment analysis. It supports multiple languages and can be easily integrated into applications via REST APIs. Since it offers pre-trained models, there's no need to manage infrastructure or train models from scratch, meeting the company's requirements.

Azure Cognitive Search focuses on indexing and searching content, not providing comprehensive NLP features.

Azure Machine Learning requires building and training custom models, which the company wants to avoid.

Azure AI Speech service handles speech recognition and synthesis, not text-based NLP tasks.

Image Classification - This is the correct answer. Image classification assigns labels to images based on the overall visual content, without pinpointing the location of specific objects. It categorizes the entire image into predefined classes for example "dog," "cat," "car" but it does not locate individual objects within the image.

Object Detection involves identifying and locating specific objects within an image, often with bounding boxes, in addition to classifying the objects. This is different from image classification, which does not detect object locations.

Optical Character Recognition (OCR) is used to extract and recognize text from images, but it is not focused on classifying images or identifying overall content.

Semantic Segmentation divides an image into regions that correspond to different object categories and labels every pixel in the image, which is more detailed than image classification. It assigns labels to every part of the image, rather than classifying the entire image at once.

Object Detection - This is the correct answer. Object detection is a computer vision technique used to identify and locate objects within an image or video. In this case, object detection would be ideal for tracking and counting individuals in the surveillance video, as it can identify and track people as they move through the store.

Image Classification - Image classification assigns a label to an image but does not provide location information about specific objects. It would not be suitable for tracking and counting individuals in real-time surveillance footage.

Optical Character Recognition (OCR) - OCR is used to extract and recognize text from images or videos. It is not applicable for tracking or counting individuals in surveillance video.

Facial Detection and Analysis - Facial detection focuses on identifying and analyzing human faces, but it would not be the best solution for tracking and counting all individuals in a store, as it is specific to detecting faces rather than general object tracking.

Azure AI Language service is specifically designed for analyzing text data using natural language processing techniques. It provides features like sentiment analysis, key phrase extraction, and entity recognition, making it ideal for extracting insights from customer feedback.

Azure AI Speech service focuses on processing spoken language, such as speech-to-text and text-to-speech conversion.

Azure Cognitive Search is used for implementing search functionalities over unstructured data.

Azure Machine Learning Studio is a platform for building and deploying custom machine learning models but doesn't offer out-of-the-box text analytics features.

Sarah should use Automated Machine Learning. This feature in Azure Machine Learning automates the process of model selection and hyperparameter tuning by experimenting with multiple algorithms and configurations to find the best model for her data. It is designed to help users who may not have deep machine learning expertise.

The Azure Machine Learning Designer allows users to build models using a drag-and-drop interface but still requires manual selection and tuning of algorithms.

Azure Data Lake Analytics is used for big data processing and query jobs, not for training machine learning models.

Azure Cognitive Services provide pre-built AI capabilities but do not allow for custom model training specific to Sarah's sales forecasting needs.

Regression - This is the correct answer. Regression is the most suitable technique for estimating the future value of investments based on historical data, market trends, and economic indicators. Regression models are used to predict continuous numerical values, making them ideal for tasks like forecasting future investment values.

Classification is used for categorizing data into predefined labels or classes (e.g., spam vs. non-spam), not for predicting continuous values like the future value of investments.

Clustering is an unsupervised learning technique used to group similar data points, but it is not suited for predicting numerical outcomes such as investment values.

Association Rule Learning is used for discovering interesting relationships or patterns in data (e.g., in market basket analysis) but is not appropriate for predicting continuous variables like investment values.

Sentiment Analysis enables the company to automatically assess the sentiment expressed in the text, categorizing it as positive, neutral or negative. This helps them understand customer attitudes and take appropriate actions.

Key phrase extraction identifies important terms or phrases, entity recognition detects named entities, and language translation converts text from one language to another; these features do not directly assess customer attitudes.

Clustering is the most suitable technique because it involves grouping similar data points together based on features, without using predefined labels. This allows the bank to identify distinct customer segments.

Regression is used for predicting continuous numerical values, which doesn't fit the goal of segmenting customers into categories.

Classification predicts categorical labels based on training data with known categories, but in this case, the categories (customer segments) are not predefined.

Reinforcement Learning involves an agent interacting with an environment to maximize cumulative reward, which is not applicable here.