Unsupervised Learning model used to cluster groups of crypto currencies based on a vast array of market data to detect best-of-class assets.

• Found best value for k Using the Original Scaled DataFrame.
• Plotted elbow curve.

• Clustered Cryptocurrencies with K-means using originaly scaled data.
• Fitted the model to make precitions based on best elbow value.
• Created a scatter plot: x-axis = price change percentage in 24 hours; y-axis = the price change percentage in 7 days.
• Colorred the points with the labels found using K-means and add the "coin_id" as a hover column.

• Optimized clusters with Principal Component Analysis.
• Plotted and overlapped both models to compare effiencies of models and elbow curves.

After visually analyzing the cluster analysis results, the impact of using fewer features to cluster the data using K-Means in a PCA model resulted in two distinct clusters that are closer to each other than the original model. Only one coin represents group 2 & 3's clusters in the PCA scatter plot.