https://github.com/fbarffmann/cryptoclustering
Clustered over 100 cryptocurrencies using K-Means and PCA to identify market patterns. Optimized clustering retained 89.5% explained variance.
https://github.com/fbarffmann/cryptoclustering
clustering crypto-analysis data-visualization hvplot k-means machine-learning pandas pca python sklearn
Last synced: 1 day ago
JSON representation
Clustered over 100 cryptocurrencies using K-Means and PCA to identify market patterns. Optimized clustering retained 89.5% explained variance.
- Host: GitHub
- URL: https://github.com/fbarffmann/cryptoclustering
- Owner: fbarffmann
- Created: 2024-09-07T16:14:58.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-04-13T16:25:06.000Z (11 days ago)
- Last Synced: 2025-04-13T17:36:06.637Z (11 days ago)
- Topics: clustering, crypto-analysis, data-visualization, hvplot, k-means, machine-learning, pandas, pca, python, sklearn
- Language: Jupyter Notebook
- Homepage:
- Size: 1.17 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Crypto Clustering with K-Means & PCA
Built a cryptocurrency clustering analysis using Python's machine learning tools. Analyzed crypto price and volume data, optimized clusters using PCA, and visualized the results to identify market patterns.
## Tools & Technologies Used
- Python
- Pandas
- Scikit-learn (K-Means Clustering & PCA)
- hvPlot
- StandardScaler
- Jupyter Notebooks
## File Structure
```text
.
├── Crypto_Clustering.ipynb # Full clustering analysis notebook
├── Resources/
│ └── crypto_market_data.csv # Cryptocurrency data
```
## Skills Demonstrated
- Clustering data with K-Means
- Finding optimal cluster count using Elbow Method
- Dimensionality reduction with PCA
- Visualizing multi-dimensional data with hvPlot
- Preparing machine learning-ready datasets
## Key Findings
- Analyzed price and volume changes for over 100 cryptocurrencies.
- Identified 4 optimal clusters using K-Means on scaled data.
- PCA optimization retained 89.5% of explained variance across 3 components.
- Visualizations showed clearer crypto market segmentation when using PCA.