Deck 7: Automated Machine Learning

Question 1

What are practitioner Elpida Ormanidou's views on the advantages of using AutoML to solve business problems?

Answer 1

No Answer

Question 2

List and explain the four key steps in the AutoML process.

Answer 2

No Answer

Question 3

AutoML, a mainly supervised approach, explores and selects models using different algorithms and compares their predictive performance.

Answer 3

True

Question 4

Identify a true statement about an unsupervised model.

A) An unsupervised model includes techniques that require a defined outcome measure.
B) An unsupervised model provides an answer key the algorithm can use to evaluate its training data accuracy.
C) Unsupervised learning enables the collection of data or a data output from a previous calculation.
D) An unsupervised model has no target variable.

Question 5

Which of the following is true of AutoML?

A) AutoML capabilities cannot be easily extended to populations beyond traditional data scientists, particularly in medium- and smaller-sized businesses.
B) AutoML is mainly an unsupervised approach that employs the traditional, manual approach of standard machine learning.
C) The AutoML platform is typically capable of analytical discovery of relationships actually present in the dataset.
D) AutoML, if implemented precisely, can replace human analytical expertise.

Question 6

Running one supervised learning model technique at a time and comparing the results with other models is a time-saving process that provides the best accuracy and prediction.

Question 7

Explain in detail the process of creating ensemble models in AutoML.

Question 8

Explain how AirBnB has been using AutoML to predict customer lifetime value for hosts and guests.

Question 9

How did Kroger, the second largest supermarket company in the world, benefit from the use of AutoML?

Question 10

Describe how Disney adopted AutoML to improve overall customer experience.

Question 11

Describe how the Philadelphia 76ers of the National Basketball Association (NBA) were able to increase ticket sales using AutoML.

Question 12

Summarize practitioner Elpida Ormanidou's views on AutoML playing a role in marketing and sales decision making.

Question 13

The boosting process in the creating ensemble models step in the AutoML process serves the purpose of

A) extracting new insights from data.
B) boosting noise and bias.
C) reducing error in the model.
D) calculating the sum of predictions from multiple models.

Question 14

Facebook uses AutoML to understand user patterns to improve business performance, such as increasing ad revenues and user engagement.

Question 15

Using AutoML in marketing analytics requires extensive coding and modeling experience.

Question 16

Which of the following is true of the supervised model of analytics?

A) In a supervised model, all data is unlabeled.
B) Association analysis and collaborative filtering are typical examples of the supervised model of analytics.
C) A supervised model is one that consists of a defined target variable.
D) A supervised model typically employs clustering methods to identify patterns in data.

Question 17

According to practitioner Elpida Ormanidou, what uses of AutoML provide the most benefit to users?

Question 18

List the questions that need to be asked when AutoML develops a predictive model.

Question 19

To achieve the most effective models while using AutoML, analysts must understand the context of the data and ensure that it meets quality standards.

Question 20

The common adage that people use when referring to ________ data is "garbage in, garbage out."

A) invalid and unreliable
B) continuous and categorical
C) complex and interrelated
D) large and diverse

Question 21

The boosting process in the AutoML process achieves error reduction by observing the error records in a model and then oversampling misclassified records in the next model created.

Question 22

Summarize practitioner Elpida Ormanidou's views on the challenges of using AutoML to solve business problems.

Question 23

The purpose of building models in the AutoML process is to

A) standardize data for a common format.
B) extract insights from data.
C) clean erroneous data.
D) reduce noise and bias.

Question 24

Identify a true statement about AutoML.

A) AutoML is mainly an unsupervised approach that employs the traditional, manual approach of standard machine learning.
B) AutoML facilitates accurate decision making for users with limited coding and modeling experience.
C) AutoML, if implemented precisely, can replace human analytical expertise.
D) AutoML capabilities cannot be easily extended to populations beyond traditional data scientists, particularly in medium- and smaller-sized businesses.

Question 25

How did Pelephone, one of the oldest and largest mobile phone providers in Israel, increase revenues by using AutoML?

Question 26

Describe how URBN, a portfolio of apparel brands, employed AutoML to improve overall customer experience.

Question 27

Describe how Sumitomo Mitsui Card Company (SMCC) in Japan applies AutoML for risk modeling and related marketing applications.

Question 28

How did Blue Health Intelligence (BHI) benefit through the use of AutoML?

Question 29

Explain the advanced ensemble methods of bagging and boosting.

Question 30

According to a recent survey conducted by Accenture, 40 percent of companies have adopted machine learning for improved sales and marketing performance.

Question 31

In the model recommendation step of the AutoML process, original data outliers and patterns are highlighted.

Question 32

The step of creating ensemble models in the AutoML process allows us to

A) handle missing data, outliers, variable selection, data transformation, and data standardization to maintain a common format.
B) provide an understanding of invisible relationships and patterns.
C) extract insight from data.
D) reduce the generalization error of the prediction.

Question 33

Which of the four key steps in the AutoML process involves handling missing data, outliers, variable selection, data standardization, and data transformation to maintain a common format?

A) preparing data
B) building models
C) recommending models
D) creating ensemble models

Question 34

List and explain several examples of AutoML applications in marketing.

Question 35

What is AutoML? Is it better than the traditional, manual approach of standard machine learning?

Question 36

An ensemble model blends the most favorable elements from all models into a single model.

Question 37

Describe how United Airlines employed AutoML to improve overall customer experience.

Question 38

Identify the correct sequence of the four key steps in the AutoML process.

A) recommending models, preparing data, building models, creating ensemble models
B) building models, creating ensemble models, recommending models, preparing data
C) creating ensemble models, preparing data, recommending models, building models
D) preparing data, building models, creating ensemble models, recommending models

Question 39

Appropriate data preparation to ensure the quality of data is an elemental first step in producing accurate model predictions.

Question 40

Explain how some companies are using AutoML tools embedded within their technology platforms to enhance business performance.