DSO 530: Decision Trees in R (Classification)

Поділитися
Вставка
  • Опубліковано 19 січ 2025

КОМЕНТАРІ • 70

  • @bashaieralyousefi4038
    @bashaieralyousefi4038 10 років тому +1

    u r an awesome teacher , u make them look very simple , thanks a lot keep up the good work , looking forward to watch more of your number 1 tutorials in R !

  • @flamboyantperson5936
    @flamboyantperson5936 8 років тому +4

    Your videos are really awesome. Please upload more on new topics.

  • @BrentBrewington
    @BrentBrewington 8 років тому

    Thanks! An RStudio tip: after partially typing the name of something (for example, a vector name), press the Tab key and it will either auto-complete or give you a list of names to complete the term

  • @vishalm2338
    @vishalm2338 8 років тому

    Awesome Tutorial. Specially I liked the Pruning of the Decision Tree part (That's bit tricky to decide)
    Thanks :-)

  • @kgangavaram
    @kgangavaram 10 років тому

    Thanks for the video ..;) its awesome Explanation , I was looking this.. Thanks for posting this .. Great work..!!

  • @thusharabandara3519
    @thusharabandara3519 2 роки тому

    Thank you very much this is a a nice and very helpful video

  • @sreenivask4407
    @sreenivask4407 10 років тому

    A very nice and lucid video.

  • @anuragmukherjee1669
    @anuragmukherjee1669 9 років тому

    Excellent piece of work !

  • @cmork2525
    @cmork2525 4 роки тому +1

    Hello, this is amazing. Kindly include the links to download that data. It is easier for us to practice with the same data. Thank you.

  • @PRAVEENSINGH7728
    @PRAVEENSINGH7728 8 років тому

    Great tutorial. Very detailed. Thanks for posting :-)

  • @ahaaha8462
    @ahaaha8462 9 років тому

    nice video, but do you need to fit the bound between high and low instead of simply assigning it to 8

  • @raminbostanabad3859
    @raminbostanabad3859 10 років тому

    Thanks. Would you please also let us know how one can add a rule (manually) to the fitted tree?

  • @parvanehshafiei1
    @parvanehshafiei1 10 років тому

    Clear with the concept..grt!

  • @borisguarisma8810
    @borisguarisma8810 11 років тому

    Excellent exercise, very clear, thank you !

  • @datascininja9401
    @datascininja9401 9 років тому

    How do I undo changes made onto the dataframe? I removed 2 variables from the Carseats dataframe, but I would like to undo that.

  • @fruofru
    @fruofru 10 років тому

    Top notch vid!
    Why is it that the rpart function builds an entirely different tree than the tree function? I could replicate the result you had with tree but not with rpart.
    Also: is there a way to display probabilities under the end nodes, under Yes/No? I couldn't find any info on this.

  • @khushi19911
    @khushi19911 9 років тому

    Great Video!! Thanks for the video!

  • @bassamal-kaaki3253
    @bassamal-kaaki3253 6 років тому

    i would like to know if there is a way that can automatically generate code when I use SPSS modeler when I am creating modules?

  • @anvikakumar3762
    @anvikakumar3762 9 років тому +2

    Heyy , Your video is a great learning source, a;though when I am trying to use it for my data it returns an error object of type 'closure' is not subsettable . I have tried to use it on the wages and error is for this line tree_model = tree[pus1$high~., training_data]. Please help me out. Thanks in advance.

  • @meghadabhade6967
    @meghadabhade6967 8 років тому

    @Abbass Al Sharif: Can you please explain how to do perturbation on decision tree? I mean how to perturb a single column with some noise and then construct decision tree.

  • @pettessy7725
    @pettessy7725 10 років тому +1

    What if we don't want to use ALL OF THE PREDICTORS to predict "HIGH". How do we go about this?

  • @rashitandon5225
    @rashitandon5225 6 років тому

    Hi Abbass, I had a question. How would you plot accuracy of tree vs tree size?

  • @prashanthiseelapu
    @prashanthiseelapu 10 років тому

    Thanks for the video, also could you tell us how to make decision trees with strings as nodes instead of numeric characters.
    Thank you .

  • @VenkateshDataScientist
    @VenkateshDataScientist 7 років тому

    Hello ,Awesome video good job.I have doubt - If the my SALES or INCOME is range from $2,000 to $70,000 then can we covert them into different group ..You mentioned only 2 groups -yes or no ----But How to handle continuous variables in R ...please reply

  • @karthip23
    @karthip23 9 років тому

    Excellent video. Thanks a lot.

  • @SachinKumaratPU
    @SachinKumaratPU 9 років тому

    Nice Video.. But I get stuck that which algorithm is it I mean C5.0, CART or it is inbuilt algorithm in R... Thanks in advance...

  • @spiderblade8696
    @spiderblade8696 9 років тому

    hello, i have a question about this, why did you erase Sales? Becouse if i let in the model, it is apear a error next

  • @aashishraina2831
    @aashishraina2831 7 років тому

    sir this material is awesome

  • @safacimenli6543
    @safacimenli6543 9 років тому

    Hello,first thanks for this video.I have a question.I apply this formula but it didn't work.I have error."'.' in formula and no 'data' argument".This is error text.Can you help me about this pleas?Thank you.

  • @mikeshinoda2806
    @mikeshinoda2806 6 років тому

    Hello. Are you running R studio on a Macbook ?

  • @dipeshsurana
    @dipeshsurana 8 років тому +1

    In Prune the tree step(13:31), how to get the best value by code, that is, not manually looking at the plot and entering the number.
    Thanks in advance.

    • @skaboy5000
      @skaboy5000 8 років тому +1

      I had the same question, and it took me a little bit to figure out how to do it. This is what did it:
      1: To find the value of the minimum CV use:
      bestprune

  • @taroice8442
    @taroice8442 11 років тому

    very nice and clear! Thanks!

  • @richardgruber3341
    @richardgruber3341 10 років тому +1

    I keep getting this error Error in type[test] : invalid subscript type 'list'
    for me my "High" is "type"
    What does this error mean and how do I fix it?

  • @garimagarg210
    @garimagarg210 10 років тому

    great video!

  • @001kanchi
    @001kanchi 8 років тому

    Thanks for awesome tutorial . Now I am stuck at very basic concepts, so can any one answer that how many input and output variable also how many terminal mode in above data.

  • @bas9049
    @bas9049 9 років тому

    Hello, What is the purpuse of the command: testing_High=High[test]? Thanks

    • @debarpitadaspal5722
      @debarpitadaspal5722 8 років тому

      I have the same question. Can anyone please explain

    • @1234sheila
      @1234sheila 8 років тому +1

      The purpose is to compare the correct result of the testing data with the result given by the decision tree. (I think :) )

  • @deveshsharma4893
    @deveshsharma4893 8 років тому

    hi , i am getting the error when im doing the cross validation .cannot coerce class ""function"" to a data.frame .
    Please help

  • @GURSIMARKAURBCE
    @GURSIMARKAURBCE 7 років тому

    Which algorithm is used here?

  • @dikshyasingh3200
    @dikshyasingh3200 2 роки тому

    Very helpful

  • @AgastyaFauzanAlfath
    @AgastyaFauzanAlfath 10 років тому

    Thanks for this video!

  • @sndrstpnv8419
    @sndrstpnv8419 8 років тому +1

    may add more info about trees with CHAID pls

  • @ashokloganathan1982
    @ashokloganathan1982 9 років тому +2

    When I try to do classification, I get the below error: Could you please help?
    Error in predict.tree(tree_model, testing_data, type = "class") :
    type "class" only for classification trees

    • @joeld23
      @joeld23 9 років тому

      Ashok Loganathan I get the same error, did you ever figure out how to fix it?

    • @icedojo
      @icedojo 9 років тому +1

      check the type of "High" object. use tranning_data$High = as.factor(tranning_data$High) and the run again.

    • @GETacclimated
      @GETacclimated 7 років тому

      You have to write "new data= testing_data"

    • @GETacclimated
      @GETacclimated 7 років тому

      newdata**

    • @ravindarmadishetty736
      @ravindarmadishetty736 6 років тому

      you have to use the keyword only predict not predict.tree

  • @huikahseng
    @huikahseng 10 років тому

    I have a question - when do we know to set.seed(1) or set.seed(2)?

    • @saku3933
      @saku3933 9 років тому

      huikahseng you can use any number as the argument. Its just easy to remember with 1,2.

    • @clairekunesh4637
      @clairekunesh4637 8 років тому

      +huikahseng Ideally you should use a random number from something like www.random.org

  • @sureshMedabala
    @sureshMedabala 10 років тому +1

    nice one.. ver clear

  • @prasad7792
    @prasad7792 8 років тому

    when High is used, i got an error ie object High is not found. can you give a solution

  • @hello20222
    @hello20222 9 років тому

    Good Job

  • @rashmimalhotra123
    @rashmimalhotra123 8 років тому

    How can we use the csv file in R ...?

  • @phoebe-phuongle2581
    @phoebe-phuongle2581 6 років тому

    what is the package?

  • @harendrakumar7647
    @harendrakumar7647 8 років тому

    nice explanation.....can i get the file you are using

  • @anooshmitadas3597
    @anooshmitadas3597 6 років тому

    Error in tree(High ~ ., training_data) :
    factor predictors must have at most 32 levels....
    i am getting this error. kindly help.

  • @sarathra46
    @sarathra46 7 років тому

    Getting error in cv.tree command. Showing as
    Error in cv.tree(fit, FUN = misclass) : not legitimate tree
    Please help on solving it

  • @ranjithkarnamsatya8400
    @ranjithkarnamsatya8400 8 років тому

    sir nice video
    i want ID3 and C4.5 implementation using R. please help me with the code

  • @yusufaltun3034
    @yusufaltun3034 8 років тому

    allah razı olsun abbass

  • @narendraprasadkarumanchi6954
    @narendraprasadkarumanchi6954 8 років тому

    > plot(tree_model)
    Error in plot.new() : figure margins too large. Please sugget how to resolve this

  • @RajeshBreaks
    @RajeshBreaks 8 років тому

    need link for carseats dataset

  • @ongtsheringlepcha6076
    @ongtsheringlepcha6076 6 років тому

    can you please provide me code

  • @steamaccount4336
    @steamaccount4336 8 років тому +1

    plot(pruned_model) : Error in xy.coords(x, y, xlabel, ylabel, log) :
    'x' is a list, but does not have components 'x' and 'y'. Getting this error. Please help