topic About Feature selection and Dimensionality reduction in SAS Data Science

About Feature selection and Dimensionality reduction

husseinmazaar — Fri, 03 Jul 2015 23:19:20 GMT

Dear Consultants,

I have dataset (599*14747) as follows: 599 observation for 6 classes and 14746 independent varaiables (predictors). the predictors are interval and the target(response) variable is nominal(categorical). I tried to find correlation to reduce the irrelevant variables using Pearson correlation but I think it finds only linear relationship only, also Pearson are changed with outliers. I need to know the optimal way to find the variables importance using statistical and data mining methods to improve the classification. notes: I have SAS Enterprise Miner 13.2

Re: About Feature selection and Dimensionality reduction

M_Maldonado — Tue, 07 Jul 2015 11:47:16 GMT

Hi Hussein,

I remember we were discussing techniques for variable selection and dimension reduction in an earlier thread ().

Do some of those techniques give you a better model than others?

Different techniques have different pros and cons. Many times you have to try a lot of them to see which one is most helpful to your data and business problem.

The course Advanced Predictive Modeling Using SAS Enterprise Miner is a good hands-on refresher of all the tools you have in your toolkit. I am considering taking that class again :smileysilly:

Post some of your findings if you have a chance.

Thanks!

-Miguel