Skip to content

cbrito3/Cryptocurrencies

Repository files navigation

Unsupervised Learning and Cryptocurrencies Report

Purpose

For this challenge about the cryptocurrency market, I am helping Martha and her firm to discover trends that will convince investors to invest in these new currencies. For this investment, a report was created as an analysis to help address what cryptocurrencies are on the trading market and how they could be grouped to create a classification system. The initial data is not ideal, so it will need to be processed to fit the machine learning models. Since there is no known output, I will use unsupervised learning. To group the cryptocurrencies, I decided on a clustering algorithm, followed by data visualizations, such as tables, 3-D scatter plot and hvplot.scatter plot, to share the findings.

As part of the report, the following will be deliver:

  • Deliverable 1: Preprocessing the Data for PCA
  • Deliverable 2: Reducing Data Dimensions Using PCA
  • Deliverable 3: Clustering Cryptocurrencies Using K-means
  • Deliverable 4: Visualizing Cryptocurrencies Results

Deliverable 1: Preprocessing the Data for PCA

Del%201%20crypto_df

Del%201%20clean%20data

Deliverable 2: Reducing Data Dimensions Using PCA

Del%202%20three%20components

Deliverable 3: Clustering Cryptocurrencies Using K-means

Del%203%20Elbow%20Curve

Del%203%20clustered_df

Deliverable 4: Visualizing Cryptocurrencies Results

Del%204%203D-Scatter

Del%204%20hvplot.table

Del%204%20new%20DataFrame

Del%204%20hvplot.scatter%20plot