https://github.com/rkschroeder/portfolio

This repository contains my portfolio of data science projects.
https://github.com/rkschroeder/portfolio

matplotlib numpy pandas sckit-learn seaborn

Last synced: 3 months ago
JSON representation

This repository contains my portfolio of data science projects.

• Host: GitHub
• URL: https://github.com/rkschroeder/portfolio
• Owner: rkschroeder
• License: apache-2.0
• Created: 2025-01-20T10:59:20.000Z (12 months ago)
• Default Branch: main
• Last Pushed: 2025-01-29T09:50:30.000Z (12 months ago)
• Last Synced: 2025-07-03T20:44:34.311Z (6 months ago)
• Topics: matplotlib, numpy, pandas, sckit-learn, seaborn
• Language: Jupyter Notebook
• Homepage:
• Size: 9.53 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Project

1. **[Customer Relationship Prediction](https://github.com/rkschroeder/Portfolio/tree/main/Customer_Relationship_Prediction)**

- Goal: Improve customer retention and increase product sales

- Task: Predict churn, appetency, and upselling to enable targeted marketing strategies

- Challenges: Heterogeneous data, imbalanced classes, and noisy data

- Action:

  - Implemented frequency encoding to handle categorical features

  - Applied an undersampling technique to address class imbalance

  - Employed `Random Forest` and `Logistic Regression` for prediction tasks

- Results: `Random Forest` outperformed `Logistic Regression` in terms of ROC AUC score for Upselling and Appetency prediction tasks, while performance for Churn prediction was similar between the two models.

- Tools: Python (`Matplotlib`, `NumPy`, `Pandas`, `Seaborn`, `scikit-learn`)