Subscribe to:
Post Comments (Atom)
My Profile
- Randall Matignon
- Randall Matignon, MS, is a CDM Systems Manager at Amgen, Inc. in South San Francisco, California. He is a SAS®/Microsoft® Office VBA programmer with over twenty years of experience as a statistical programmer and applications developer in the pharmaceutical heathcare, and biotechnology industries, and he has a broad knowledge of several programming languages including SAS®, VBA Microsoft Office, PL-SQL, S-Plus®, and Java.
5 comments:
Hi,
SAS Enterprise Miner streamlines the data mining process to create highly accurate predictive and descriptive models based on analysis of vast amounts of data from across the enterprise.It Supports the entire data mining process with a broad set of tools.
spa soil screening
Hello,
I'm a novice to Data Mining and SAS Enterprise miner. I'm trying to create a lift chart using predicted probabilities of response. I have splitted the validation data using the Splitting rules derived from Training data but my Lift/Response numbers are not matching with that of Work.LiftData numbers. I suspect I'm not ranking the predicted probabilities correctly. Could you please tell me how SAS EM computes deciles?
Thanks
Srikanth
This book is very good and useful to novices to Data Mining and SAS Enterprise miner.
Hi,I am a novice to Data Mining and SAS Enterprise mine too.Recently i study this book carefully and find some nodes are diffculte to understand.For example ,how does the princomp/dmneuarl node work? how to write the regression eqution from the output the node? Could you please offer an example and elaborate on it for me?Thanks.
Hello,
I found this blog about data mining and would like to ask you a few questions.
1) I have 10 different input variables in my dataset, do I have to choose the most important or can I use all of then at the moment of using a tree, neural and regression model?
2) How can I interpretate the results of regression and reural model? They are difficult to see the patterns.
3) Validation and training should be the lowest possible, or the similar possible?
4) After applied 3 models (tree, regression and neural) how can I know which one is the best?
Thanks
Post a Comment