Pages: [1]
  Print  
Author Topic: ARIMAX and Optimization Operator for both feature selection and parameters  (Read 1539 times)
knnbayaz
Newbie
*
Posts: 9


« on: November 20, 2012, 10:30:59 PM »

Hi

ARIMAX (Auto Regressive Integrated Moving Average with Exogeneous Input) algorithm will be very good i think. It is ver usefull  for time series forecasting. And optimization algorithm can handle both feature selection and parameter optimization. Optimization algorithm optimize feature selection and parameter optimization at the same time. Maybe this is available in rapidminer but i don' know if there is
Logged
Marius
Administrator
Hero Member
*****
Posts: 1794



WWW
« Reply #1 on: November 21, 2012, 12:25:13 PM »

As far as I know ARIMAX is not yet implemented for RapidMiner. I will propose it as part of the rework of our series extension, but I can't give you any time estimates or even say if it will be accepted at all.

Best regards,
Marius
Logged

Please add [SOLVED] to the topic title when your problem has been solved! (do so by editing the first post in the thread and modifying the title)
Please click here before posting.
knnbayaz
Newbie
*
Posts: 9


« Reply #2 on: November 22, 2012, 04:36:31 PM »

Thanks Marius

I want to know if rapidminer team are planning an optimization tool which optimize feature selection and parameter optimization concurrently. The article in the data mining literature show us this kind of tools are very succesful. To my knowledge, rapidminer can not optimize feature selection and parameter optimization concurrently.
Logged
Marius
Administrator
Hero Member
*****
Posts: 1794



WWW
« Reply #3 on: November 23, 2012, 11:22:54 AM »

Heya,

you can simply put the feature optimization operator inside of the parameter optimization operator or vice versa. That way you get a cascade which optimizes features and parameters at the same time.

Best, Marius
Logged

Please add [SOLVED] to the topic title when your problem has been solved! (do so by editing the first post in the thread and modifying the title)
Please click here before posting.
Pages: [1]
  Print  
 
Jump to: