COURSE 2: MICROSOFT AZURE MACHINE LEARNING FOR DATA SCIENTIST

Module 4: Create A Clustering Model With Azure Ai

MICROSOFT AZURE DATA SCIENTIST ASSOCIATE (DP-100) PROFESSIONAL CERTIFICATE

Complete Coursera Study Guide

Enroll in Coursera Microsoft Azure Data Scientist Associate (DP-100)

Last updated:

TABLE OF CONTENT

INTRODUCTION – Create A Clustering Model With Azure Ai

Clustering is an unsupervised machine learning technique that groups similar entities based on their features, revealing patterns and structures within datasets without predefined labels. In this module, you will learn to create clustering models using the Azure Machine Learning designer, which provides an intuitive drag-and-drop interface for model development and deployment.

You will explore various clustering algorithms, gaining practical experience in selecting and fine-tuning models for different data types. By the end of this module, you will understand how to leverage Azure Machine Learning designer to build effective clustering models, uncovering valuable insights in your data.

PRACTICE QUIZ: KNOWLEDGE CHECK

1. True or False?

Clustering is a form of machine learning that is used to group similar items into clusters based on their predictions.

  • True 
  • False (CORRECT)

Correct: Clustering is a form of machine learning that is used to group similar items into clusters based on their features.

2. What type of machine learning technique does clustering examples?

  • Reinforcement 
  • Supervised
  • Unsupervised (CORRECT)

Correct: Clustering is an example of unsupervised machine learning.

3. You create a machine learning experiment based on a clustering model. Now you want to use the model in an inference pipeline. Which module should you use to infer cluster predictions from the model?

  • Train clustering model
  • Assign data to clusters (CORRECT)
  • Score model

Correct: Assign Data to Clusters module generates cluster predictions from a trained clustering model.

4. When using a clustering module, what algorithm let’s you group items into a number of clusters you specify?

  • C-Means algorithm
  • B: L-Means algorithm
  • K-Means algorithm (CORRECT)

Correct: The K-Means algorithm groups items into the number of clusters you specify.

5. Suppose you are testing a K-Means clustering model. If you would want your model to assign items to one of four clusters, which parameter/property should you configure on the module?

  • Set random seed to 4
  • Set stratified split
  • Set number of centroids to 4 (CORRECT)

Correct: The number of centroids defines the number of clusters.

QUIZ: TEST PREP

1. Which of the following is a clustering algorithm?

  • K-Means (CORRECT)
  • Two-Class Neural Network
  • Two-Class Logistic Regression

Correct: K-Means is a clustering algorithm.

2. What is the purpose of a clustering model?

  • Makes forecasts by estimating the relationship between values
  • Answers simple two-choice questions
  • Separates similar data points into intuitive groups (CORRECT)

Correct: Clustering models have the purpose of separating similar data points into intuitive groups.

3. Which of the following scenarios can be resolved by applying clustering modules/algorithms?

Select all that apply.

  • A bike rental company that wants to predict the number of customers for the next day so that it will assure the necessary staff and cycles.
  • A radio company that wants to apply tags (like rock, pop, R&B etc) to songs or artists. (CORRECT)
  • A social media company that wants to group similar users based on their posts. (CORRECT)

Correct: Clustering models have the purpose of separating similar data points into intuitive groups.

Correct: Clustering models have the purpose of separating similar data points into intuitive groups.

4. When evaluating a clustering model, what metrics can you visualize in the Evaluate results section?

Select all that apply.

  • Average distance to cluster center (CORRECT)
  • Maximal distance to cluster center (CORRECT)
  • Average distance to other center (CORRECT)
  • Number of points (CORRECT)

Correct: The metrics that can be visualized in the Evaluate results section of a clustering module are: Average distance to other center, Average distance to cluster center, Number of points, Maximal distance to cluster center.

Correct: The metrics that can be visualized in the Evaluate results section of a clustering module are: Average distance to other center, Average distance to cluster center, Number of points, Maximal distance to cluster center.

Correct: The metrics that can be visualized in the Evaluate results section of a clustering module are: Average distance to other center, Average distance to cluster center, Number of points, Maximal distance to cluster center.

Correct: The metrics that can be visualized in the Evaluate results section of a clustering module are: Average distance to other center, Average distance to cluster center, Number of points, Maximal distance to cluster center.

5. You are building an Azure Machine learning pipeline that involves a clustering module. You need to prepare the data and change some of the numeric values from the dataset to use a common scale, without distorting differences in the ranges of values or losing information. 

Which module should you apply?

  • Normalize Data (CORRECT)
  • Split data
  • Edit metadata

Correct: The goal of normalization is to change the values of numeric columns in the dataset to use a common scale, without distorting differences in the ranges of values or losing information.

6. True or False?

Clustering is an example of supervised machine learning, in which you train a model to separate items into clusters based purely on their characteristics or features.

  • True
  • False (CORRECT)

Correct: Clustering is an example of unsupervised machine learning, in which you train a model to separate items into clusters based purely on their characteristics or features.

7. A Hospital Care chain wants to open a series of Emergency-Care wards within a region. The chain knows the location of all the maximum accident-prone areas in the region. They have to decide the number of the Emergency Units to be opened and the location of these Emergency Units, so that all the accident-prone areas are covered in the vicinity of these Emergency Units. 

Which type of machine learning model is best to be applied in this scenario?

  • Regression 
  • Classification
  • Clustering (CORRECT)

Correct: Clustering models have the purpose of separating similar data points into intuitive groups. With clustering, the chain can group accidents by emergency units.

8. You want to train a model where there is no previously known cluster value (or label) from which to train the model. 

Which type of machine learning would you use?

  • Unsupervised machine learning (CORRECT)
  • Supervised machine learning

Correct: Clustering is an example of unsupervised machine learning, in which you train a model to separate items into clusters based purely on their characteristics, or features. There is no previously known cluster value (or label) from which to train the model.

CONCLUSION – Create A Clustering Model With Azure Ai

In conclusion, clustering is a vital unsupervised machine learning technique for grouping similar entities based on their features, uncovering patterns and structures in unlabeled data. Throughout this module, you have learned to create and optimize clustering models using the intuitive Azure Machine Learning designer. With practical experience in selecting and fine-tuning various algorithms, you are now equipped to effectively leverage this tool to build robust clustering models, enabling you to derive valuable insights and patterns from your data with confidence.

Previous Module
Weekly Breakdown

Quiztudy Top Courses

Popular in Coursera

Liking our content? Then, don’t forget to ad us to your BOOKMARKS so you can find us easily!

Expanded Thumbnail Promo1

Mood Zone for Studying & Relaxing