Practice Quiz 2 - confusing question & feedback statement

“Question 2: A data pipeline is a series of data processing steps such as:…”

‘Data Analysis’ is marked as incorrect, which confuses me. But the feedback is even more confusing:

“Let’s recheck that. The predictive signal is extracted from the data through feature selection.”

I don’t see the connection between the question which is on data processing in a pipeline and the feedback which is talking about predictive signals and feature selection. Are you saying that Data Analysis is all about feature selection and predictive analytics and nothing else? I thought feature selection was quite a specific activity. Also, I’m not sure why feature selection should necessarily be excluded from a data pipeline.

Agree. I had the same complaint while going thru that quiz.

Hi,
I have the same doubt.
I would also like to highlight the CD foundation MLOps reference architecture that was shown in the previous lectures. Here, the pipeline does contain data analysis (1st stage Data prep, third component).


In my understanding, Data analysis is not defined/limited to feature selection rather an inspection that can also include statistics like distribution of data based on classes or categories which could further help in cleaning, deciding preprocessing steps, and train/test split.
Here, it seems to be used in a different way.