Inspired from Udacity Data Scientist Nanodegree
The idea behind this project is to perform customer segmentation based on historical data provided by Arvato Financial Solutions.
We need to analyze a "CUSTOMERS" dataset and figure out how customers are similar to or differ from the general population at large ("AZDIAS" dataset). Then, using information from that analysis, we need to make predictions on users who were the target of a marketing campaign ("MAILOUT" dataset).
Software dependencies:
All sofware dependencies are stated in the notebooks, however, the main libraries used are:
- Pandas-profiling
- Numpy
- Pandas
- Matplotlib
- Seaborn
- Scikit-Learn
- XG-Boost
Disclaimer: the dataset is private and can not be shared as stated by Udacity's rules. I personally took it from other people repositories which provided it under MIT License.