Microsoft Azure AI Fundamentals Microsoft Azure AI Fundamentals MCQ 3 — Questions and Answers
Question 1: Which metric would be most appropriate for evaluating a binary classification model predicting rare fraud events?
- Accuracy
- Mean Squared Error
- Precision-Recall AUC (Correct answer)
- R-squared
Correct answer: Precision-Recall AUC
For imbalanced datasets like fraud detection, Precision-Recall AUC is more informative than accuracy, which can be misleadingly high when the positive class is rare.
Question 2: Azure Cognitive Services Language Understanding (LUIS) is primarily used for:
- Translating text between languages
- Extracting intents and entities from natural language input (Correct answer)
- Generating human-like text responses
- Converting speech to text
Correct answer: Extracting intents and entities from natural language input
LUIS analyzes natural language to identify the user's intent and extract relevant entities, enabling conversational AI applications.
Question 3: Which Azure service is specifically designed to build, train, and deploy machine learning models at scale?
- Azure Cognitive Services
- Azure Machine Learning (Correct answer)
- Azure Data Factory
- Azure Logic Apps
Correct answer: Azure Machine Learning
Azure Machine Learning is a cloud platform designed to support the full ML lifecycle from data preparation to model training, deployment, and monitoring.
Question 4: What is a 'confusion matrix' used for in machine learning?
- Visualizing neural network architecture
- Evaluating classification model performance by showing true vs predicted labels (Correct answer)
- Displaying feature correlations in a dataset
- Comparing training and validation loss curves
Correct answer: Evaluating classification model performance by showing true vs predicted labels
A confusion matrix shows the counts of true positives, true negatives, false positives, and false negatives, providing a comprehensive view of classifier performance.
Question 5: Which Azure AI capability can identify the specific objects present in an image along with their locations?
- Image classification
- Object detection (Correct answer)
- Optical character recognition
- Semantic segmentation
Correct answer: Object detection
Object detection not only identifies what objects are in an image but also provides bounding boxes showing where each object is located.
Question 6: In the context of Azure AI, what is a 'knowledge base' in Azure QnA Maker (now part of Azure AI Language)?
- A database of ML model weights
- A structured collection of question-and-answer pairs used by a chatbot (Correct answer)
- A repository of training images
- A set of pre-trained language models
Correct answer: A structured collection of question-and-answer pairs used by a chatbot
A knowledge base in QnA Maker is a curated set of Q&A pairs that enables a chatbot to answer user questions by matching them to stored answers.
Question 7: Which principle of responsible AI focuses on ensuring AI systems work correctly and safely across various conditions?
- Fairness
- Inclusiveness
- Reliability and safety (Correct answer)
- Accountability
Correct answer: Reliability and safety
Reliability and safety ensures AI systems perform as designed under normal and adverse conditions without causing unintended harm.
Microsoft Azure AI Fundamentals Microsoft Azure AI Fundamentals MCQ 3
Which metric would be most appropriate for evaluating a binary classification model predicting rare fraud events?