CertBlaze

Microsoft

AI-900 Free Practice Questions — Page 8

Question 64

HOTSPOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Each correct selection is worth one point.

Illustration for AI-900 question 64
Show Answer
Correct Answer: No Yes Yes
Explanation:
1) from=it means the input must be Italian, not English. 2) from=en with to=fr and to=it translates English into French and Italian. 3) The Translator service supports document translation, including English to French.

Question 65

HOTSPOT - Select the answer that correctly completes the sentence.

Illustration for AI-900 question 65
Show Answer
Correct Answer: Object detection
Explanation:
Object detection identifies and locates multiple types of objects within a single image, unlike image classification (single label), image description (text summary), or OCR (text extraction).

Question 66

You plan to develop a bot that will enable users to query a knowledge base by using natural language processing. Which two services should you include in the solution? Each correct answer presents part of the solution. NOTE: Each correct selection is worth one point.

A. Language Service
B. Azure Bot Service
C. Form Recognizer
D. Anomaly Detector
Show Answer
Correct Answer: A, B
Explanation:
To build a bot that lets users query a knowledge base using natural language, you need NLP capabilities and a bot framework. The Language Service provides natural language understanding and question answering over a knowledge base, while Azure Bot Service hosts and manages the bot that handles user interactions. Form Recognizer and Anomaly Detector are unrelated to conversational knowledge querying.

Question 68

You have a dataset that contains the columns shown in the following table. You have a machine learning model that predicts the value of ColumnE based on the other numeric columns. Which type of model is this?

A. analysis
B. clustering
C. regression
Show Answer
Correct Answer: C
Explanation:
The model predicts a numeric target (ColumnE) from other numeric features, which is the definition of a regression problem. Clustering is unsupervised and does not predict a specific target variable, and 'analysis' is not a machine learning model type.

Question 69

HOTSPOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Each correct selection is worth one point.

Illustration for AI-900 question 69
Show Answer
Correct Answer: Yes Yes No
Explanation:
Clustering is an unsupervised technique that groups items based on similarity without predefined labels. Organizing documents or patients into groups based on features fits clustering. Predicting severity levels (mild, moderate, severe) is classification, not clustering.

Question 70

You need to create a customer support solution to help customers access information. The solution must support email, phone, and live chat channels. Which type of Al solution should you use?

A. machine learning
B. computer vision
C. chatbot
D. natural language processing (NLP)
Show Answer
Correct Answer: C
Explanation:
A chatbot is the appropriate AI solution for customer support across email, phone, and live chat because it is designed to interact with users conversationally across multiple channels. Chatbots can integrate natural language processing internally, but the solution type that delivers the end-to-end support experience is a chatbot.

Question 71

You are building a chatbot that will use natural language processing (NLP) to perform the following actions based on the text input of a user. • Accept customer orders. • Retrieve support documents. • Retrieve order status updates. Which type of NLP should you use?

A. sentiment analysis
B. named entity recognition
C. translation
D. language modeling
Show Answer
Correct Answer: D
Explanation:
The chatbot must understand user intent expressed in free-form language (place an order, retrieve documents, check status) and generate appropriate actions and responses. Language modeling underpins intent understanding and response generation across all these tasks. Named entity recognition is useful as a supporting technique to extract details like product names or order numbers, but by itself it cannot determine the user’s intended action.

Question 72

HOTSPOT - For each of the following statements, select Yes if the statement is true. Otherwise, select No. NOTE: Each correct selection is worth one point.

Illustration for AI-900 question 72
Show Answer
Correct Answer: No Yes No
Explanation:
Question answering does not directly query Azure SQL databases. Question answering is designed to return consistent answers from a knowledge base for similar questions. Determining user intent is handled by conversational language understanding (LUIS/CLU), not question answering.

Question 73

You need to track multiple versions of a model that was trained by using Azure Machine Learning. What should you do?

A. Explain the model.
B. Register the model.
C. Register the training data.
D. Provision an inference cluster.
Show Answer
Correct Answer: B
Explanation:
In Azure Machine Learning, registering a model stores it in the workspace with versioning metadata. Each time you register a model with the same name, a new version is created, allowing you to track, manage, and compare multiple versions of the trained model over time.

Question 74

You are developing a system to predict the prices of insurance for drivers in the United Kingdom. You need to minimize bias in the system. What should you do?

A. Remove information about protected characteristics from the data before sampling.
B. Take a training sample that is representative of the population in the United Kingdom.
C. Create a training dataset that uses data from global insurers.
D. Take a completely random training sample.
Show Answer
Correct Answer: B
Explanation:
To minimize bias, the training data should accurately reflect the population for which predictions are made. A training sample representative of the UK driving population ensures different demographics, driving behaviors, and risk profiles are proportionally included, reducing systematic bias. Removing protected characteristics alone does not eliminate indirect bias, global data may introduce irrelevant differences, and a purely random sample may still be unrepresentative if the source data is skewed.

$19

Get all 230 questions with detailed answers and explanations

  • Instant download HTML + PDF delivered the moment payment clears.
  • Secure Stripe checkout we never see or store your card details.
  • 7-day refund if files are defective see our refund policy.

Full AI-900 study guide →