The goal of this project is to predict which individuals are most likely to convert into becoming customers for a mail-order sales company in Germany.
- A Descriptive Jupyter Notebook
Results and discussion were published on Medium: Creat a Customer Segmentation for Arvato Financial Services
In this blog, we dive into a real life machine learning project provided by Arvato Financial Solutions, a Bertelsmann subsidiary:
- Investigated Demographics data of general population of Germany and data for customers of a mail-order company.
- Preprocessed the dataset based on column/feature property.
- Apply Unsupervised Learning Algorithms, namely PCA and KMeans to segment the population (into different clusters) to recommend the potential customers for the company.
- Took a deeper look at two main clusters and compare them by checking the differences of several randomly choose features.
- Apply Supervised Learning to predict whether or not a person became a customer of the company following the campaign.
- Investigate the most important feature trained by machine learning model and compare the feature distribution between target/non-target population.