This project aims to create a deep learning model to predict if a customer would continue to deposit money to expand a bank's revenue. For this objective, the bank carried out a campaign to collect customers' details as well as their needs and satisfactions. The data collected can be retrieved in the train.csv file in the repo.
This model achieved a 90% accuracy in determining the customers' deposit commitment to the bank.
A sneak peek of the model developed and model report are as below:
👉 Train.csv
👉 customer_segmentation_model.py
👉 ModulesCustomer.py
👉 model.h5
👉 pickle_files folder that contains encoded features
👉 photos folder which contains the following images:
- classification report
- epoch accuracy and epoch loss (from tensorboard)
- model accuracy and model loss (plotted in python)
- model architecture
- This project is done using Python 3.8 on Spyder. This project used the following modules:
-
The sample datasets and model has already been included in the repository.
-
You may download all the necessary files (dataset & python files) to run the project on your device.
This dataset is taken from: HackerEarth HackerLive: Customer Segmentation | Kaggle