During the process of Machine Learning, when should you review evaluation metrics.
A. Before you train a model.
B. After you clean the data.
C. Before you choose the type of model.
D. After you test a model on the validation data.
Show Answer
Correct Answer: D
Explanation: Evaluation metrics are used to assess how well a trained model performs on unseen data. Therefore, they are reviewed after the model has been trained and then tested on validation (or test) data. Before training, cleaning data, or choosing a model type, there are no predictions to evaluate yet, so metrics cannot be meaningfully reviewed.
Question 96
You need to make the written press releases of your company available in a range of languages.
Which service should you use?
A. Speech
B. Language
C. Translator
D. Personalizer
Show Answer
Correct Answer: C
Explanation: The requirement is to translate written press releases into multiple languages. The Translator service is designed specifically for text translation between languages, unlike Speech (audio), Language (analysis like sentiment), or Personalizer (recommendations).
Question 97
You have a website that includes customer reviews.
You need to store the reviews in English and present the reviews to users in their respective language by recognizing each user's geographical location.
Which type of natural language processing workload should you use?
A. key phrase extraction
B. speech recognition
C. language modeling
D. translation
Show Answer
Correct Answer: D
Explanation: The requirement is to store reviews in English and present them to users in their own languages based on location. This requires converting text from one language to another, which is a translation workload. Key phrase extraction, speech recognition, and language modeling do not perform language-to-language conversion.
Question 99
HOTSPOT
-
Select the answer that correctly completes the sentence.
Show Answer
Correct Answer: Custom Vision
Explanation: Custom Vision allows you to upload and label your own images to train custom object detection models. Computer Vision provides only prebuilt models, while Form Recognizer and Video Analyzer are for documents and video analysis respectively.
Question 100
You have a natural language processing (NLP) model that was created by using data obtained without permission.
Which Microsoft principle for responsible AI does this breach?
A. reliability and safety
B. privacy and security
C. inclusiveness
D. transparency
Show Answer
Correct Answer: B
Explanation: Using data obtained without permission violates Microsoft’s Responsible AI principle of privacy and security, which requires that data be collected, accessed, and used lawfully and with proper consent, while protecting individuals’ rights and information.
Question 101
HOTSPOT -
Select the answer that correctly completes the sentence.
Show Answer
Correct Answer: analysis
Explanation: The system evaluates qualities of the face in the image—such as exposure, noise, and occlusion—to give feedback. This goes beyond simply locating a face (detection) and does not involve identifying who the person is (recognition), which makes it facial analysis.
Question 102
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.
Show Answer
Correct Answer: Yes
Yes
Yes
Explanation: Object detection locates objects with bounding boxes, can detect multiple instances of the same object, and can be trained to recognize multiple object classes (e.g., different types of damaged products).
Question 103
In a machine learning model, the data that is used as inputs are called ________.
Select the answer that correctly completes the sentence.
A. dataset
B. labels
C. variables
Show Answer
Correct Answer: C
Explanation: In machine learning, the inputs provided to a model are called variables (more precisely, input variables or features). A dataset is the overall collection of data, which can include inputs, labels, and other fields, while labels refer to the target outputs the model learns to predict. Therefore, the correct completion is "variables."
Question 104
DRAG DROP -
Match the services to the appropriate descriptions.
To answer, drag the appropriate service from the column on the left to its description on the right. Each service may be used once, more than once, or not at all. You may need to drag the split bar between panes or scroll to view content.
NOTE: Each correct selection is worth one point.
Show Answer
Correct Answer: Language Service
Speech
Explanation: The Language Service (formerly QnA Maker) enables natural language queries against a knowledge base. The Speech service provides real-time speech-to-text transcription.
Question 106
You have insurance claim reports that are stored as text.
You need to extract key terms from the reports to generate summaries.
Which type of AI workload should you use?
A. anomaly detection
B. natural language processing
C. computer vision
D. knowledge mining
Show Answer
Correct Answer: B
Explanation: Extracting key terms and generating summaries from text documents involves understanding and processing human language, which is a core task of natural language processing (NLP). Anomaly detection, computer vision, and knowledge mining do not directly address text understanding and summarization.
$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.
