Applied understanding of the K-means algorithm and principal component analysis (PCA) to classify cryptocurrencies according to their price fluctuations across various timeframes
Applies understanding of the K-means algorithm and principal component analysis (PCA) to classify cryptocurrencies according to their price flucuations across various timeframes and specifically, examines price changes over 24 hours, 7 days, 30 days, 60 days, 200 days, and 1 year. Uses PCA to determine which data attribute have the greatest influence on component analysis outcomes.
The primary skills demonstrated include the ability to:
- Use the standardScaler() module from scikit-learn to normalize CSV files
- Optimize data using principal component analysis (PCA)
- Find best k-value using scaled data and PCA data
- Cluster data using K-Means
- Find best k-value using PCA data
- Determine weighing of principal components
Python Libraries
- pandas
- KMeans (from sklearn.cluster)
- PCA (from sklearn.decomposition)
- StandardScalar (from sklearn.preprocessing)
Jamie Bond | [email protected] | Connect on Linkedin
Adapted from instructions starter files provided by The Artificial Intelligence Boot Camp at UNC Charlotte, including:
- Data: crypto_market_data.csv
- Instructions: Crypto_clustering_starter_code.ipynb