Arvato Financial Services Customer Segmentation Report

This project aims to analyze the demographics data of a mail-order sales company in Germany and compare it against the general population demographics. The goal is to create a customer segmentation report using unsupervised learning techniques and identify the characteristics of the company's core customer base. Additionally, the project involves predicting potential customers for a marketing campaign using a separate dataset.

Project Overview

The project is based on real-life data provided by Bertelsmann Arvato Analytics. It involves four datasets:

1. Udacity_AZDIAS_052018.csv: Demographics data for the general population of Germany.
2. Udacity_CUSTOMERS_052018.csv: Demographics data for the company's customers.
3. Udacity_MAILOUT_052018_TRAIN.csv: Demographics data for individuals targeted in a marketing campaign (train set).
4. Udacity_MAILOUT_052018_TEST.csv: Demographics data for individuals targeted in a marketing campaign (test set).

The project consists of two main parts:

Part 1: Customer Segmentation Report

In this part, the project analyzes the relationship between the demographics of the company's customers and the general population. The steps involved are as follows:

1. Data Preprocessing: The datasets are loaded and preprocessed to handle missing values, convert data types, and drop irrelevant columns.
2. Dimensionality Reduction: Principal Component Analysis (PCA) is used to reduce the dimensionality of the datasets while retaining important information.
3. Clustering: K-means clustering algorithm is applied to the PCA-transformed datasets to group similar data points together. The optimal number of clusters is determined using the elbow method.
4. Comparison: The cluster labels of the customer dataset and the general population dataset are compared to identify similarities and differences, providing insights into the characteristics of the company's customer base.

Part 2: Predictive Modeling

In this part, the project uses the insights gained from Part 1 to predict potential customers for the marketing campaign. The steps involved are as follows:

1. Data Preprocessing: The train dataset is preprocessed in a similar manner as in Part 1.
2. Feature Selection: Features from the train dataset that are most relevant for predicting customer conversion are identified.
3. Model Training and Evaluation: A machine learning model is trained on the selected features and evaluated using the test dataset. The model predicts which individuals are most likely to become customers for the company.

Repository Contents

• Data folder: Contains the four datasets provided for the project.
• Arvato Project Workbook.ipynb: Jupyter Notebook containing the code and analysis for the project.
• README.md: Readme file providing an overview of the project and its contents.

Dependencies

The following libraries are used in the project:

• numpy
• pandas
• matplotlib
• seaborn
• sklearn

Please ensure that these libraries are installed before running the code.

How to Use

To run the project, follow these steps:

1. Clone the repository to your local machine.
2. Install the required dependencies mentioned above.
3. Open the Arvato Project Workbook.ipynb file using Jupyter Notebook or any other compatible environment.
4. Run the notebook cells sequentially to perform the data analysis, segmentation, and predictive modeling tasks.

Note: The code assumes that the data files are located in the Data folder within the project directory. Please make sure to update the file paths if the data files are stored elsewhere.

Acknowledgment:

We would like to express our sincere gratitude to Udacity for their mentorship and guidance, and to Arvato Financial Services for providing the datasets that made this project possible. We also extend our appreciation to the open-source community for their contributions in developing the libraries and frameworks that played a crucial role in our analysis.

Blog Post

You can find a detailed report on this project following this post on Medium: https://medium.com/@p.dehghani88/customer-segmentation-report-for-arvato-financial-services-5dacc0d8166b