/E-Commerce_UK_Retailer_ML

Machine Learning Project for UK Retailer: Customer-Segmentation, E-Commerce-Sales-Forecast and Customer-Purchasing-Patterns

Primary LanguageJupyter Notebook

Machine Learning Project for UK Retailer: Customer-Segmentation and Customer-Purchasing-Patterns

Thanks to an e-commerce actual data transactions from an UK retailer in this notebook we are going to explore an e-commerce dataset transactions from an UK retailer, this dataset lists purchases made by approximately 40000 customers through a period of time of one year (from 12/01/2010 to 12/09/2011).

The main aim of this notebook is to develop a machine learning model that allows to anticipate the purhcases that will be made by a new customer, over the next year according to its firsts purchases.

The notebook was developed by performing the next steps:

  • Data Cleaning.
  • Feature Exploration.
  • Understanding Product Categories.
  • Customers Categories.
  • Classifying Customers.
  • Testing Predictions.
  • Explaining The Decissions of The Model.

Say Hello to KModes for Categorical Clustering

A brief introduction to an algorithm for categorical clustering is thaught in the process of developing an accurate customer segmentation, this algorithm can be used by installing the library:

pip install kmodes