BootCamp Week 18. Unsupervised Machine Learning
Cryptocurrencies is a trending market and its very easy to get confused among all the different options that the market have. In this project, we were asked to create a report about all the cryptocurrencies are on the trading market and how could be grouped in a more understandable to a potential clients. Since we don´t have a known output, I will use Unsupervised Machine Learning to group the cryptocurrencies using a clustering K-means algorithm and data visualization to show the results.
The steps to process this data were:
1. Preprocess the data to get it able to ML Analysis.
2. Reducing data dimensions using PCA (Personal Component Analysis).
3. Clustering data using K-means Algorithm
4. Visualizing Results.
From the original 1,253 cryptocurrencies in the original data file, after the filtering process we worked with only 532 cryptos. Then I reduced data dimensions to three principal components and I created a new DataFrame. In the next deliverable I plot an Elbow Curve using to find the best value for clustering groups. After that,in order to prepare the data for the clustering process.
