Describe fundamental principles of machine learning on Azure for Microsoft Azure AI Fundamentals (AI-900)

Sample Practice Questions

1. Question 1

   For a machine learning progress, how should you split data for training and evaluation?
   1. A. Use features for training and labels for evaluation.
   2. B. Randomly split the data into rows for training and rows for evaluation.
   3. C. Use labels for training and features for evaluation.
   4. D. Randomly split the data into columns for training and columns for evaluation.
   Explanation

   The correct answer is: B. Randomly split the data into rows for training and rows for evaluation..

   You split data along rows because each row represents one independent observation, and you want both the training set and the evaluation set to contain complete observations drawn from the same overall distribution. Random row-wise selection produces representative subsets and avoids biases that fixed orderings could introduce. Using features for training and labels for evaluation is incorrect because the model needs both features and the matching label in training so it can learn the mapping, and evaluation also requires features paired with true labels so predictions can be compared. Swapping that and using labels for training and features for evaluation is even less coherent because a model cannot learn without features. Splitting by columns separates features from each other, which destroys the meaning of an observation and makes training impossible. Random row-wise splitting is the correct approach.

2. Question 2

   HOTSPOT - You are developing a model to predict events by using classification. You have a confusion matrix for the model scored on test data as shown in the following exhibit. Use the drop-down menus to select the answer choice that completes each statement based on the information presented in the graphic. NOTE: Each correct selection is worth one point. Hot Area:
   Explanation

   Box 1: 11 - TP = True Positive. The class labels in the training set can take on only two possible values, which we usually refer to as positive or negative. The positive and negative instances that a classifier predicts correctly are called true positives (TP) and true negatives (TN), respectively. Similarly, the incorrectly classified instances are called false positives (FP) and false negatives (FN). Box 2: 1,033 - FN = False Negative - Reference: https://docs.microsoft.com/en-us/azure/machine-learning/studio/evaluate-model-performance

3. Question 3

   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. Hot Area:
   Explanation

   Anomaly detection encompasses many important tasks in machine learning: Identifying transactions that are potentially fraudulent. Learning patterns that indicate that a network intrusion has occurred. Finding abnormal clusters of patients. Checking values entered into a system. Reference: https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/anomaly-detection

4. Question 4

   HOTSPOT - To complete the sentence, select the appropriate option in the answer area. Hot Area:
   Explanation

   Reference: https://docs.microsoft.com/en-us/azure/machine-learning/team-data-science-process/create-features

5. Question 5

   DRAG DROP - You plan to deploy an Azure Machine Learning model as a service that will be used by client applications. Which three processes should you perform in sequence before you deploy the model? To answer, move the appropriate processes from the list of processes to the answer area and arrange them in the correct order. Select and Place:
   Explanation

   Reference: https://docs.microsoft.com/en-us/azure/machine-learning/concept-ml-pipelines

Other Microsoft Azure AI Fundamentals (AI-900) domains

• Describe Artificial Intelligence workloads and considerations (76 questions)
• Describe features of computer vision workloads on Azure (46 questions)
• Describe features of Natural Language Processing (NLP) workloads on Azure (66 questions)