Unsupervised machine learning, PCA, KMeans, etc.
Python, K-Means, scikit-Learn, Pandas, Plotly Express, hvPlot
Accountability Accounting is interested in offering a new cryptocurrencies investment portfolio for its customers. The company, is seeking a report of what cryptocurrencies are on the trading market and how cryptocurrencies could be grouped toward creating a classification for developing a new investment product.
Created an unsupervised machine learning model to analyze data on the cryptocurrencies traded on the market. The provided data set was processed with this Python code and prepared for Principle Component Analysis to reduce data set dimensions to three. Cryptocurrency clusters were predicted using the K-means algorithm with an elbow-curve technique to select the best K-value.
Finally, the cryptcurrency data was visualized in a number of ways. The code plotted the data with each principle component on an axis in a 3D hvPlot. Then it created a filterable table of cryptocurrencies. Lastly, the data was visualized with a logarithmic 2D chart comparing coin supply to coins mined.