| dc.description.abstract | One of the biggest data domain and most demanding use cases of recent time is Customer
churn prediction. For a healthy and growing business churn prediction is an important
indicator. This project aims to develop a churn prediction for banking sector. For
predicting customer churn I have chosen hyper parameters of deep learning. I have
collected a dataset from kaggle, which have 10000 rows and 14 columns. I divided the
dataset into two parts. One is train data which have contains 75% data and another is test
data which contains 25% data of the whole dataset. I have done some analysis on the data
set. I have used deep learning hyper parameter. Using deep learning the trained model
gives 79% accuracy. I have also used some machine learning algorithm such as Random
Forest, Decision Tree, K-nearest neighbor (KNN) and Logistic regression. Among this
four algorithm Random Forest has given better accuracy which is about 85%. | en_US |
