AI - 900 (Activity)

What is the primary source from which machines learn in the realm of machine learning?

• Algorithms

• Algorithms

• Data

• Programming Language

Answer explanation

Solution: Data

Rationale: Machine learning fundamentally operates on data. As mentioned in the provided text, machines learn "from data." Algorithms process this data to extract patterns and relationships, but without data, the algorithms have nothing to learn from.

What does the Anomaly Detector service in Microsoft Azure primarily offer?

• A cloud storage solution

• A machine learning model for wildflower classification

• A platform for designing machine learning solutions without code

• An API for creating anomaly detection solutions

Answer explanation

Solution: An API for creating anomaly detection solutions

Rationale: According to the provided text, the Anomaly Detector service in Microsoft Azure provides an application programming interface (API) for developers to build anomaly detection solutions. The focus is specifically on detecting anomalies, rather than storage or any other machine learning tasks.

What does the task of semantic segmentation in computer vision primarily involve?

• Reading and extracting text from images.

• Classifying images based on their overall contents.

• Identifying and labeling individual objects within an image.

• Classifying individual pixels in the image according to the object they belong to.

Answer explanation

Solution: Classifying individual pixels in the image according to the object they belong to.

Rationale: As outlined in the provided information, semantic segmentation is an advanced machine learning technique where individual pixels in an image are classified according to the object to which they belong. Unlike object detection, which identifies and labels individual objects, semantic segmentation deals with individual pixels.

Which service in Microsoft Azure is primarily used to translate text between different languages?

• Azure AI Vision

• Azure AI Speech

• Azure AI Language

• Azure AI Translator

Answer explanation

Solution: Azure AI Translator

Rationale: The provided text specifically mentions that the Azure AI Translator service is used to translate text between more than 60 languages. While other Azure AI services deal with various language and speech tasks, Azure AI Translator is the one specifically focused on translation between languages.

Which of the following best describes the main function of Azure Cognitive Search in the context of knowledge mining?

• It provides a framework for building AI systems responsibly.

• It's primarily used to ensure privacy and security of AI systems.

• It is a private, enterprise search solution for building indexes and making unstructured data searchable.

• It assists in testing AI applications for safety and reliability.

Answer explanation

Solution: It is a private, enterprise search solution for building indexes and making unstructured data searchable.

Rationale: The provided text specifically mentions that Azure Cognitive Search is a private, enterprise, search solution. Its main function is to build indexes which can be used for internal or public-facing assets. By utilizing AI capabilities such as image processing and natural language processing, Azure Cognitive Search can index previously unsearchable documents and extract insights. This helps transform large volumes of unstructured data into a searchable knowledge store, which is at the heart of knowledge mining.

What is the primary purpose of machine learning?

• To create fancy charts

• To predict unknown values using mathematical and statistical models

• To generate large volumes of data

• To automate mundane tasks

Answer explanation

Solution: To predict unknown values using mathematical and statistical models

Rationale: Machine learning uses mathematical and statistical techniques to create models that can predict unknown values, as mentioned in the initial explanation.

Which of the following is NOT a type of supervised machine learning task?

• Regression

• Clustering

• Classification

• Time series forecasting

Answer explanation

Solution: Clustering

Rationale: Clustering is an unsupervised machine learning approach where data is grouped into similar clusters. Supervised learning requires known label values, while clustering does not.

In Azure Machine Learning, what is the main use of an Azure Container Instance (ACI)?

• To deploy a service for long-term production scenarios

• To manage the compute targets for data science activities

• To deploy a service suitable for testing

• To automatically select and tune machine learning algorithms

Answer explanation

Solution: To deploy a service suitable for testing

Rationale: Azure Container Instances (ACI) provides a suitable deployment target for testing services in Azure Machine Learning. For production scenarios, it's recommended to use Azure Kubernetes Service (AKS).

Which evaluation metric is used to calculate the square difference between predicted and actual values, where smaller values indicate more accurate predictions?

• Mean Absolute Error

• Normalized Root Mean Squared Error (NRMSE)

• R-squared value

• Accuracy

Answer explanation

Solution: Normalized Root Mean Squared Error (NRMSE)

Rationale: NRMSE is calculated by squaring the errors across all test cases, finding the mean of these squares, and then taking the square root. It standardizes the RMSE metric, so it can be used for comparison between models. Smaller values of NRMSE indicate more accurate models.

An automobile dealership wants to use historic car sales data to train a machine learning model. The model should predict the price of a pre-owned car based on its make, model, engine size, and mileage. What kind of machine learning model should the dealership use automated machine learning to create?

• Classification

• Regression

• Time series forecasting

Answer explanation

The answer is Regression. To predict a numeric value that is continuous, use a regression model.