Infrared spectrum data treatment before processing

grandoniagrandonia Member Posts: 9 Contributor II
edited November 2018 in Help
Dears, we have been working with unscrambler for IR spectra (where we preteat the spectra with first and second derivatives and smoothing before doing anything else with the data... is there a way do to the same with RM?

Here is a picture of what I would like to do (go from A to C and then start the DM  ;))

image

And... at unscrambler I normally don't discretize the label column, use the numeric values. At RM only working nice creating cathegories (low, medium, high), which is fine too, but... wonder if I could do it the otherwise around too, or is this computationally too intensive.

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist
    Dear grandonia,

    sorry i am unsure what you want to do. Simply the conversion A->C or some data mining on C?

    ~Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • grandoniagrandonia Member Posts: 9 Contributor II
    Hi Martin,

    Simply the conversion from A->C inside Rapidminer :)
  • grandoniagrandonia Member Posts: 9 Contributor II

    Hi Martin, any news to make a smoothing + 2nd derivative of the data? With this procedure I can get rid of many variables that are useless :) Thanks!!

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,503 RM Data Scientist

    Hi,

     

    there are moving average and a differentiate operator in the series extension. Have you checked them?

     

    Best,

    Martin

    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.