topic Re: Please help to resolve this query in SAS Data Science

Question about SAS Enterprise Miner fact sheet

rghv — Thu, 05 Jul 2018 18:21:12 GMT

What is the meaning of this line. It was mentioned in the SAS documentation in SAS Enterprise Miner 5.3 for Desktop

Mine the selected principal components using predictive modeling techniques.

Re: Please help to resolve this query

PaigeMiller — Thu, 05 Jul 2018 13:38:07 GMT

Please give us the context to this statement (perhaps the entire paragraph or several paragraphs). Or better yet give us a link to the documentation where you see this.

Re: Please help to resolve this query

rghv — Thu, 05 Jul 2018 13:43:44 GMT

Hi,

The line is mentioned n the below link.

http://www.sas.com/technologies/analytics/datamining/miner/factsheet_desktop.pdf

Its in page 6 under heading Dimension reduction.

Thanks.

Re: Please help to resolve this query

PaigeMiller — Thu, 05 Jul 2018 13:58:23 GMT

First, this is not the documentation for Enterprise Miner, as you stated originally. It is marketing hype, it's gobbledygook, it's meaningless.

Or maybe it means you can use any predictive modeling technique available in Enterprise Miner on the principal components. Which really narrows things down, doesn't it?

Re: Please help to resolve this query

ChrisHemedinger — Thu, 05 Jul 2018 14:55:37 GMT

That is a really old fact sheet. Mentions support for Windows XP!

For more recent SAS Enterprise Miner info, see the Learn page for the product.

Re: Please help to resolve this query

Reeza — Thu, 05 Jul 2018 15:09:56 GMT

This is a product feature list.

Mine the selected principal components using predictive modeling techniques.

Likely means you can take the principal components and analyze them using other procedures - they can go into regression models or you can track them over time to see if the factors are changing.

@rghv wrote:

What is the meaning of this line. It was mentioned in the SAS documentation in SAS Enterprise Miner 5.3 for Desktop

Mine the selected principal components using predictive modeling techniques.