Using unsupervised machine learning to classify a portfolio of currently tradable cryptocurrencies
- crypto_data.csv
- Jupyter Notebook
- Libraries: pandas, numpy, hvplot, plotly, sklearn
1 - The crypto_data.csv
dataset was cleaned and preprocessed in order to perform PCA (Principal Component Analysis).
2 - The resulting dataframe was reduced to 3 dimensions using PCA.
3 - K-Means analysis was performed using the PCA data to determine the number of clusters (4) and create a prediction model.
4 - Visualizations were created to show:
- K-means clustering in 3D
- a table sortable by feature
- a scatterplot of coin supply vs coins mined
The completed code crypto_clustering.ipynb
is available here