Creating a cryptocurrency investment report that includes what cryptocurrencies are on the trading market and how they could be grouped to create a classification system using an unsupervised machine learning model.
Helping a hypothetical investment bank offer a crypto-focused portfolio for its customers.
Part 1: Preprocessing the Data for PCA
-
read in file
-
remove columns with unnecessary data for unsupervised learning
-
get rid of nulls
-
remove untradable coins
Part 2: Reducing Data Dimensions Using PCA
-
assigning the new data to 3 different components
-
create a PCA DataFrame
Part 3: Clustering Cryptocurrencies Using K-means
-
use KMeans to cluster the crypto data
-
print predictions made by the model
-add predictions to original data
- create an elbow curve
Part 4: Visualizing Cryptocurrencies Results
-
create a 3D and 2D scatter plot
-
count all the currently available cryptos