Isolation Forest: A Tree based approach for Outlier Detection (Clearly Explained)

Поділитися
Вставка
  • Опубліковано 1 січ 2025

КОМЕНТАРІ • 13

  • @HafeezUllah
    @HafeezUllah Місяць тому

    Very nice explanation.

  • @aadhyatiwari9688
    @aadhyatiwari9688 6 місяців тому +1

    One of the best explanations out here! Thankyou

  • @roi_4_dayz
    @roi_4_dayz Місяць тому

    Very nice explanation. Thanks for the help!

  • @machinelearningplus
    @machinelearningplus  Рік тому

    Want to learn more ML? Checkout edu.machinelearningplus.com/s/pages/ds-career-path
    - Become fundamentally strong in Data Science and ML!

  • @ankithahj1694
    @ankithahj1694 Місяць тому

    Very good explanation.
    Suppose for single set a feature values, I am getting as outlier. How to find out top features contributed for the decision?

  • @VikasVerma-xf6hb
    @VikasVerma-xf6hb 9 місяців тому +1

    Nice ...Thanks

  • @prathipmathavan3089
    @prathipmathavan3089 7 місяців тому

    helped me a lot, keep up he good work!

  • @azogdevil
    @azogdevil 3 місяці тому

    Thanks 🙏

  • @ashrithadepu
    @ashrithadepu Рік тому

    Hi, my dataset consists of categorical values and I’ve label encoded them to use isolation forest model. But how to evaluate my model? What metrics should I follow?

    • @machinelearningplus
      @machinelearningplus  Рік тому

      If your 'features' are categorical, don't label encode then. Label encoding is meant for Target variables.
      Evaluating models can be done as you would with any other predictive model

    • @ashrithadepu
      @ashrithadepu Рік тому +1

      @@machinelearningplus but I don’t have a target variable and all the data is categorical how do you think I can proceed?? Btw thanks for your reply

    • @r0cketRacoon
      @r0cketRacoon 5 місяців тому

      @@ashrithadepu one hot encoding, that'all

  • @mikeclark4611
    @mikeclark4611 Рік тому

    🙂 Promo sm