📊 Hierarchical Clustering Analysis in R

Personal project by Wiricardo for practicing hierarchical clustering techniques and customer segmentation in R.
This repository contains a comprehensive analysis comparing agglomerative and divisive clustering methods on customer personality data using tidymodels and factoextra.

---

Project Structure

Clustering-R/
├── clustering-code-R.qmd    # Main analysis document (Quarto)
└── README.md

---

🎯 Analysis Overview

Dataset: Customer Personality Analysis
Methods: Agglomerative (AGNES) and Divisive (DIANA) hierarchical clustering
Variables: Total spending, age, and total purchase quantity
Result: 3 customer segments identified

---

📈 Key Findings

• Divisive clustering (0.42 silhouette) outperformed agglomerative (0.35 silhouette)
• Cluster 1: High-value mature customers (avg. age 64 years, spending $1,094)
• Cluster 2: Low-value occasional buyers (avg. spending $147)
• Cluster 3: Premium young buyers (avg. age 44 years, spending $1,238)

---

🚀 Technologies

• R & Quarto - Reproducible analysis
• tidyverse & tidymodels - Data manipulation and modeling
• cluster - AGNES and DIANA algorithms
• factoextra - Dendrograms and silhouette plots
• plotly - Interactive 3D visualizations

---

Data Source

• Customer Personality Analysis Dataset - Demographic data, purchase history, and marketing campaign responses

---

License

This project is licensed under the MIT License.
See the LICENSE file for more details.

---

Contact

For questions or collaboration:
GitHub profile → wiricardo