Naviotech_Project6_Bank_Customer_Churn_Prediction_Project

This is the famous Bank Customer Churn Prediction Project which is based on predicting whether a customer will remain or churn in a bank using the Random Forest Classification Algorithm.

The Random Forest Classification Algorithm is a very useful algorith as it gives high accuracy and also makes the predictions based on the Decision Trees. This model basically tells a bank that which customers are likely to remain in the bank and which customers are likely to churn or leave from the bank. 1 represents a customer who left the bank whereas 0 represents the customer who remain in the bank.

Here, the predictions of the customer churn has been made on mainly 9 parameters which are as:

Total Charges

Monthly Charges

Tenure

Contract Type

Internet Service

Payment Method

Tech Support

Online Security

Paperless Billing

Model Accuracy : 77.19 %

Here, for further visualization, the several plots have been made which are as:

   Correlation Heatmap 
   Confusion Matrix 
   Histograms 
   Boxplots   

The Python libraries used include:

         Streamlit 
         Numpy 
         Pandas 
         Matplotlib 
         Seaborn 
         Scikit-Learn 
         

Here, along with the analysis on the Jupyter Notebook, I've also made the Streamlit UI App to make it more accessible for the users.

Streamlit UI Link : https://cmangla-bank-churn-predictor.streamlit.app/