Discuss Databricks Databricks Machine Learning Associate Exam Topic 2 Question 45

In which of the following situations is it preferable to impute missing feature values with their median value over the mean value?

AWhen the features are of the categorical type

BWhen the features are of the boolean type

CWhen the features contain a lot of extreme outliers

DWhen the features contain no outliers

EWhen the features contain no missing no values

Suggested Answer: C

Imputing missing values with the median is often preferred over the mean in scenarios where the data contains a lot of extreme outliers. The median is a more robust measure of central tendency in such cases, as it is not as heavily influenced by outliers as the mean. Using the median ensures that the imputed values are more representative of the typical data point, thus preserving the integrity of the dataset's distribution. The other options are not specifically relevant to the question of handling outliers in numerical data. Reference:

Data Imputation Techniques (Dealing with Outliers).

Get Premium Databricks Machine Learning Associate Questions as Interactive Web-Based Practice Test or PDF Real Exam Format Detailed Explanations Instant Access Get Your Practice Test

12 comments · newest first

Charisse 2 months ago

I’m not sure. What about A? Categorical features need modes, right?

Trinidad 2 months ago

Agreed, C makes sense. Median is more robust.

Avery 3 months ago

I think C is the best choice. Outliers can skew the mean.

Franklyn 3 months ago

A and B don’t make sense for imputation at all.

Nicolette 3 months ago

Median is great for outliers, totally agree with C!

Edgar 3 months ago

Wait, why would you use median for categorical features?

Eleonora 3 months ago

I disagree, mean can be better in some cases.

Lorriane 3 months ago

C is definitely the right choice!

Cordelia 4 months ago

I feel like the median is generally a safer choice for skewed distributions, but I’m not confident about the other options.

Zona 4 months ago

I practiced a similar question, and I think the median is preferable when the data has outliers, which makes me lean towards option C.

Blondell 4 months ago

I’m not entirely sure, but I think imputing with the mean can skew the data if there are outliers, so maybe C is the right choice?

Chuck 4 months ago

I remember discussing that the median is less affected by outliers, so I think it might be better when there are extreme outliers.

Databricks Machine Learning Associate Exam - Topic 2 Question 45 Discussion