Decision tree

desaijayanti
desaijayanti New Altair Community Member
edited November 2024 in Community Q&A
Hello! We are trying to create a Decision tree for a sample data using RapidMiner. The data contains 10 columns. However, the decision tree is categorising based on only one column. Is there any change in the settings which has to be done? If so can someone please let us know? Thanks in advance!!
Tagged:

Best Answer

  • RalfKlinkenberg
    RalfKlinkenberg New Altair Community Member
    Answer ✓
    Deactivating pruning and pre-pruning can help.
    Setting the "minimal gain" parameter of the decision tree learner to a smaller value, e.g. 0.05 or 0.01 usually helps as well.

Answers

  • Marco_Boeck
    Marco_Boeck New Altair Community Member
    Hi,

    can you please post your process XML ("View" -> "Show View" -> "XML")? Ideally you also include a link to your data. My guess would be that "minimal gain" or "min leaf size" or "minimal size for split" parameters might be responsible.

    Regards,
    Marco
  • RalfKlinkenberg
    RalfKlinkenberg New Altair Community Member
    Answer ✓
    Deactivating pruning and pre-pruning can help.
    Setting the "minimal gain" parameter of the decision tree learner to a smaller value, e.g. 0.05 or 0.01 usually helps as well.