What are good reasons to develop a nave Bayes classifier model?

Handles very high dimensional data and resistant to over-fitting

Handles correlated variables and handles missing variables

Handles categorical variables and handles numeric variables

Well calibrated and easy to implement

Handles very high dimensional data and resistant to over-fitting

What is a benefit of Spark in-memory data processing as opposed to using MapReduce?

Avoids writing intermediate data to disk, which speeds up processing

Supports processing unstructured data, which MapReduce does not allow

Removes the need to use disks at all, which reduces cost

Allows parallel processing, which MapReduce does not support

When building a K-means clustering model, you notice that the clusters did not segment on variables that you expected. What should you do?

Check that the data was properly scaled

Multiply each variable by its standard deviation

Check that the data was properly scaled

In time series analysis, what statement describes a MA(q) process?

Current deviation from the time series mean depends on the q previous deviations

Current deviation from the time series mean depends on the quotient q

Current time series value depends on the q previous values

Current time series value depends on the fitted polynomial of order q

After running a density plot you realize that the data has a long tail to the right. What can you do to make the dataset more normally distributed?

Apply a logarithmic transformation

Use a scatter plot to obtain a better picture

Use a histogram to obtain a better picture

Apply a logarithmic transformation

What action occurs during feature selection in the model building phase of the data analytics lifecycle?

Identify the most useful input variables

Create new combinations of attributes

Overfit the model to improve prediction accuracy

Identify the most useful input variables

Select a superset of variables to shorten training times

What are three built-in data types in the R programming language?

Boolean, integer, and character

You build a decision tree to classify five different types of customers based on their browsing history from a sample of 500. The resulting decision tree has 17 layers. One of the leaf nodes has only three customers.What do you conclude?

Due to large number of layers, there may be an overfitting problem

The decision tree needs to be rebuilt without the three customers

The decision tree needs to be rebuilt to see if the results change

The sample size is too small, so the classes may not be accurate

Due to large number of layers, there may be an overfitting problem

What is the primary role of a business intelligence analyst on an analytics project?

Provides business-domain expertise

Extracts business data from source systems

Ensures business milestones are met

Provides business-domain expertise

Defines business goals for the analytics project

What is part of the model output for a linear regression?

Coefficients indicating relative impact of the input variables on the outcome

The assignment of each input datum to a cluster

Coefficients indicating relative impact of the input variables on the outcome

The set of all rules X -> Y with minimum support and confidence

Probability score for each possible class label

DELL D-DS-FN-23 Practice Test 2 - Free Online Exam Prep & Questions

In addition to quantitative and technical skills, what is a key aspect of the profile of a data scientist?

Become a Premium Member for full access

Unlock Premium Member

DELL D-DS-FN-23 Practice Test 2

Question

DELL D-DS-FN-23 Practice Tests

Practice Tests