The purpose of this project is to use unsupervised machine learning to analyze a database of cryptocurrencies and create a report including the traded cryptocurrencies classified by group according to their features. This classification report could be used by an investment bank to propose a new cryptocurrency investment portfolio to its clients. I have used the following methods for the analysis:

• Preprocessed data for the database,
• Reduced the data dimension using Principal Component Analysis,
• Used clustering cryptocurrencies with K-Means,
• Visualization of classification results with 2D and 3D scatter plots.

Data Source: crypto_data.csv, Software: Python, Jupyter Notebook

Following the preprocessing and cleaning stage I had a total of 532 tradable cryptocurrencies.

I didn't know what would be the output of the analysis so I was using unsupervised machine learning to identify clusters of the cryptocurrencies. I populated the elbow curve below using the K-Means method iterating on k values from 1 to 10.

The best k value appears to be 4 so I would conclude on an output of 4 clusters to categorize the crytocurrencies.

This 3-D scatter plot was obtained using the PCA algorithm to reduce the crytocurrencies dimensions to three principal components. scatter plots show the distribution and the four clusters of cryptocurrencies.

Most of the cryptocurrencies are part of class #0 and #1.

Plotting the scatter plot from two cryptocurrency features does not efficiently segregate the different classes. Utilized the PCA algorithm, as it is the right method for better visualizations.

I have identified the classification of 532 cryptocurrencies based on similarities of their features. Particularities of each group need to be analyzed to determine their performance and potential interest for the investment bank's clients.