/cltv_prediction_with_bgnbd_gamma

Primary LanguageJupyter NotebookApache License 2.0Apache-2.0

CLTV Prediction with BG-NBD and Gamma-Gamma Models

FLO, a shoe store, wants to define a roadmap for their sales and marketing actions. For the company to make medium-long term plans, the potential value that existing customers will bring should be predicted.

Dataset Summary

The dataset consists of information obtained from the past shopping behavior of customers who made their last purchase as Omnichannel (both online and offline purchases) from FLO in 2020 - 2021.

Variables

Variable Definition
master_id: Unique customer number.
order_channel: The channel that the purchase was made. (Android, iOS, Desktop, Mobile)
last_order_channel: The channel that the last purchase was made.
first_order_date : First purchase date of the customer.
last_order_date: Last purchase date of the customer.
last_order_date_online: Last online purchase date of the customer.
last_order_date_offline: Last offline purchase date of the customer.
order_num_total_ever_online: Total online purchase count of the customer.
order_num_total_ever_offline: Total offline purchase count of the customer.
customer_value_total_ever_offline: Total value spent on offline purchases by the customer.
customer_value_total_ever_online: Total value spent on online purchases by the customer.
interested_in_categories_12: The categories that customer purchased from in the last 12 months.

Run on your PC

Clone the project

  git clone https://github.com/cagkangrsy/cltv_prediction_with_bgnbd_gamma

Go to project directory

  cd cltv_prediction_with_bgnbd_gamma

Run the notebook.

Data Protection

The dataset is not published in the project due to Data Protection Laws.