Business Analytics Ch. 7

Automated Machine Learning

(AutoML) is a process that automates the task of developing analytical models with machine learning.

Arrange the key steps in the Automated Machine Learning (AutoML) process in the correct order of occurrence. (Place the first step at the top.)

1. Preparing data 2. Building models 3. Creating ensemble 4. Suggesting models

Which of the following points should be kept in mind when marketers use Automated Machine Learning (AutoML)?

The AutoML platform is only able to analytically discover relationships that actually are present in a dataset.

When using Automated Machine Learning (AutoML), marketers should keep in mind that _____.

the implementation of AutoML does not replace human expertise

What are the main types of analytics methods? (Check all that apply.)

A supervised model An unsupervised model

Match the types of analytics methods (in the left column) with their descriptions (in the right column).

A supervised model It consists of a defined target variable. An unsupervised model It does not have a target variable.

Which of the following is most likely part of the "model recommendation" stage of the Automated Machine Learning (AutoML) process?

Determining accuracy by monitoring how well a model identifies relationships and patterns in a dataset and uses this knowledge to predict the result

In the "model recommendation" stage of the Automated Machine Learning (AutoML) process, higher levels of a model's accuracy are measured in the original datasets used to develop the model.

False

What are the steps involved in the advanced ensemble method called bagging in the Automated Machine Learning (AutoML) process? (Check all that apply.)

Generating several random smaller samples from a larger sample Executing a model on each sample and combining the outcomes

ensemble model

In the Automated Machine Learning (AutoML) process, the best approach is to combine various algorithms, blending information from more than one model into a single "super model." This type of model is referred to as a(n) ___ ___.

Identify a true statement about the advanced ensemble method called boosting in the Automated Machine Learning (AutoML) process.

It aims at improving a model's performance and reducing misclassification.

Which of the following statements is true about Automated Machine Learning (AutoML)?

It enables accurate decision making for users with limited coding and modeling experience.

Identify the true statements about Automated Machine Learning (AutoML). (Check all that apply.)

It has helped overcome the shortage in advanced data analytics skills. It explores and selects analytical models with the help of different algorithms and compares their predictive performance.

In the context of the Automated Machine Learning (AutoML) process, identify the true statements about an ensemble model. (Check all that apply.)

It has the most favorable elements from all models. It reduces problems such as noise and bias.

In the context of model building in the Automated Machine Learning (AutoML) process, identify a true statement about models.

It is used to gain insights from data.

Which of the following are the key steps in the Automated Machine Learning (AutoML) process? (Check all that apply.)

Preparing data Creating models and then building ensemble models

Which of the following is most likely part of data preparation in the Automated Machine Learning (AutoML) process?

Selecting variables

data preparation

Successful use of Automated Machine Learning (AutoML) begins with _____.

Boosting

Which of the following is the advanced ensemble method in the Automated Machine Learning (AutoML) process that observes the error records in a model and then oversamples misclassified records in the next model created?

In the advanced ensemble method called bagging in the Automated Machine Learning (AutoML) process, observations in a sample can be copied and placed in a second or even a third sample. This process is called _____.

bootstrap sampling

Automated Machine Learning (AutoML) _____.

has become more commonly adopted in marketing analytics

During model building in the Automated Machine Learning (AutoML) process, tens of models are usually created automatically after an analyst _____.

states the dependent variable of interest