RapidMiner Wisdom Banner

ID3 algorithm in rapidminer

fatimidveilfatimidveil Member Posts: 45 Contributor II
edited January 11 in Help
hi everyone , i am performing id3 algorithm on my 15 attributes but when ide tree is generated it consist of only 8 attributes what should i do know ? why this happen . .i am using cross validation 
id3 algorithm didn't select the 7 attribute in tree .
am waiting for your help

Tagged:
Jasmine_

Answers

  • IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,715  RM Founder
    That is actually quite normal that not all attributes are used.  The whole point of a decision tree is to figure out which are the most helpful features and in which order.  Sometimes not all are needed for a classifier and sometimes adding more may even make the model generalize less well.
    Hope this helps,
    Ingo
    RapidMiner Wisdom 2020
    February 11th and 12th 2020 in Boston, MA, USA

    sgenzerJasmine_
  • fatimidveilfatimidveil Member Posts: 45 Contributor II
    okay are can i use those attribute that are discarded  by algrithm   in my final result ?
    because some attribute seem important in my result but discarded by decison tree 


    Jasmine_
  • IngoRMIngoRM Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University Professor Posts: 1,715  RM Founder
    Sorry, can you please rephrase a bit?  Not sure if I got you here.  What other result are you referring to?
    Cheers,
    Ingo
    RapidMiner Wisdom 2020
    February 11th and 12th 2020 in Boston, MA, USA

    Jasmine_
  • fatimidveilfatimidveil Member Posts: 45 Contributor II
    My query is: can i use those attritube that are not selected by tree in my final result discussion .
    Becaz some attritube are important for my final result but algorithm discard them .waiting for your help and response

    Jasmine_
  • varunm1varunm1 Moderator, Member Posts: 999   Unicorn
    Can i use those attribute that are discarded  by algrithm   in my final result ?
    I don't think it is possible to add attributes into an id3 based decision tree.


    Jasmine_
  • fatimidveilfatimidveil Member Posts: 45 Contributor II
    My query is: can i use those attritube that are not selected by tree in my final result discussion .
    Becaz some attritube are important for my final result but algorithm discard them .waiting for your help and response
    Jasmine_
  • fatimidveilfatimidveil Member Posts: 45 Contributor II
    Can i make different id3 algorithm for same data using different attritubes that are important for us ..
    Jasmine_
  • MillaMilla Member Posts: 2 Contributor I
    edited 2:32AM
    That is quite ordinary that not all characteristics are utilized. The general purpose of a choice tree is to make sense of which are the most supportive highlights and in which request. Here and there not all are required for a classifier and once in a while including more may even make the model sum up less well.
    Burger king survey
    Jasmine_
  • fatimidveilfatimidveil Member Posts: 45 Contributor II
    Thankyou guys for your help ..😊
    Jasmine_
Sign In or Register to comment.