Bank customer churn prediction using Machine Learning

Поделиться
HTML-код
  • Опубликовано: 28 окт 2024

Комментарии • 8

  • @emsif
    @emsif Месяц назад

    does the .corrwith method don´t use as default the pearson correlation? But you use it also for categorical features like gender. Further the target is also categorical. For this case it´s more appropriate to use Kendall and Chi2.

  • @channelfisikaasik1124
    @channelfisikaasik1124 6 месяцев назад +2

    Hey! thanks for the tutorial but I want to ask a question. First of all in this video you don't explain about the plotting between each data. I have plotted the same data as you and the result is that the churn and not churn is overlapping for every plot whether it is age and balance or balance and estimated salary and etc so can you explain how does randomforest(which is a training of many decision trees)can get a good result in overlapping churn and not churn because if the churn and not churn are overlapping each other then the parameter used to predict data will not accurate because it is not certain what features is good enough to predict the churn because the separation is not clear
    thanks!!

  • @ziadsamer5246
    @ziadsamer5246 Год назад

    Very Nice explanation

  • @chukwuebukamichaelezeokeke6844
    @chukwuebukamichaelezeokeke6844 8 месяцев назад

    what happened to france ?

  • @smitvidja8388
    @smitvidja8388 Год назад

    can you provide github repository of this or google colab code file ?

    • @smitvidja8388
      @smitvidja8388 Год назад

      you can share link here or in description so other can also use it.

    • @how_to_404
      @how_to_404  Год назад

      Sorry for the delayed response. PFA the link. github.com/asishrj91/Customer_Churn_Prediction_Model.git

    • @smitvidja8388
      @smitvidja8388 Год назад +1

      Thank you for providing the link@@how_to_404