Rapidminer Video Tutorial – Genetic Feature Selection

Posted on by

Note: This video has been superseded by the new Rapidminer 5.0 Video Tutorials (see videos #3 and #4).

This video has been updated 7/20/08. I know its been a while but I finally got off my butt and finished the next installment of my RapidMiner Video Tutorial 3. This one is about using the data preprocessing operator called Genetic Feature Selection and I talk about how you can use it to build a trading model.

The video link is to a flash (FLV) file. You must have a stand alone FLV player to download and view this tutorial video.

  • Digital Dude

    Tom,

    Very nice!!! Another great tutorial dude ;-)

    Now this is why I come here 8^)

    Cordially,

    -Digital Dude-

  • Tom

    DD: Thanks! There are more to come!

  • jw

    not sure if anyone else had this problem, but about 3/4 through the tutorial, the sound stops working. tried it at work and at home and got same result. any ideas?

    thanks for the tutorials

  • Tom

    jw: I placed two stop arrows in the lower right corner of the window. Everything stops until you press them to continue. Have you tried that?

  • Tom

    jw: I see the problem, when I get to my work computer I’ll fix it and reload.

  • jw

    thanks and appreciate the videos.

  • Maricela

    Hello I’m begginer in this about rapidminer, I have some doubts and I hope you can help me.
    My problem is that I don’t know how to make a linearRegression, I now that maybe it’s so simple but I don’t have any idea…I try to do it but there are mistakes :(
    Thank you for your help

  • Tom

    Maricela: Are you using the LinearRegression operator?

  • Mary

    Hi! I just want to ask if it’s possible for me to group the data using clustering, then through Rapidminer, I could select one of the groups then apply a neural net..

    It would really help! Thanks!

  • Tom

    Mary: Yes, you can do that if you like.

  • Édouard

    Hi Tom.

    Thanks for the material you provide on your blog … despite few problems (not yet issued videos, sound cuttings … ). ;-)

    Somewhere along this video (“Rapidminer Video Tutorial – Genetic Feature Selection”), you point out that your genetic feature selection model in RapidMiner is quite different from ‘classic’ time-series (econometric) modelling. Can you tell a little but more on that ???

    Where online can I find the calculations behind the operators in RapidMiners (the learners, the preprocessing- and postprocessing-operators, …) ????

    Thanks.
    Édouard, France

  • Tom

    Edouard: I’m glad you like them. RM has a set of time series operators that allows you to do the classic trend analysis, I do things a bit differently. That’s what I was referring too in the video.

    Most of the operators were built from Weka operators. Which are open source. You can find information on the RM operators at the RM website.

  • jayaraj

    hi,
    I am working on building a model for insurance fraud detection. I am planning to use predictive modeling technique in order to detect the fraud claims. Can i use this rapid miner for my problem and how?…..thanks in advance……

    regards,
    jayaraj

  • Elmo

    I watched you tutorial on rapidminer it was a great.
    when i tried to do the samething with my own data on Excel i got error messages:
    ” Parameter ‘excel_file’ is not set and has no default value.”

    would you please tell me what did I forget to do or most probably did not know that I have to do to my excel sheet?

    thank you

  • Jai Majeti

    Hiii,
    We have a problem in dividing a data set into two different sets Training and test sets using rapidminer so that we can run different algorithms like ID3, decission tree etc in rapid miner. We tried dividing it using simple validation using stratified sampling, we got a setup error. Can any body diagonize that problem for us???
    Thank you
    Jai

  • http://www.neuralmarkettrends.com Tom

    Jai,
    Sometimes when you get an error in the Validation operator its because the data isn’t fed from the data loader operator correctly. Have you correctly set up your data loader operator? Are the Label and ID inputs correctly set?

    Also, try sampling using linear or shuffled and let the decision tree operator break your data into like groups.

  • http://www.neuralmarkettrends.com Tom

    Elmo, can you send me a screenshot of it?

  • Koosha

    Hi Tom,

    I need some help with rapidminer. I saw your tutorials and thought you might be able to help with this.

    I need to search a dataset to find similar trends for a given pattern/trend.
    A given pattern is available in a spreadsheet (for example a trend of heart rate values) and a separate spreadsheet represents a history of values for the same attribute (for example heart rate of a patient during 90 days). The goal is to find patterns similar to the given trend.

    I am wondering about the process and operators that I can use.

    I should mention this is not a prediction problem and a simple process using Euclidean distance (to match the input pattern with window moving along the testset) can do the job.

    Thanks,
    – Koosha