Crypto Punks EDA in R
The skills the authors demonstrated here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.
Background:
- Crypto Punks launched in June 2017 by Larva Labs (John Watkinson and Matt Hall), bought over by Yuga Labs
- Collection of 10,000 unique 24x24 pixel images
- Algorithm generated
- Early non-fungible tokens (NFT)
Objective:
- Explore CryptoPunks attributes
- What are the attributes?
- Which attributes are common? Rare?
Visualizations:
Conclusion:
- Having seven or zero attributes is most rare, and three attributes is most common
- Hair attribute is most common
- Teeth attribute is most rare
- These attributes make the CryptoPunks unique - one of a kind - and can be used to access their rarity and value
Recommendations:
- Collection items and speculation investments
- Use attributes to find Punk that you like
- Look at past transaction prices, bids, and offers
- Use a combination of attributes, price history, and niche demand for value
- Go for Aliens, zero attributes, special hair attributes (beanie, top hat, wild white hair)
Next Steps:
- Link transaction history to available Punks to find relationship between attributes and price
- Price and demand change over time
- Feature engineer to create a “Rarity” feature - estimating the rarity of Punks
- Natural Language Processing of sentiment on CryptoPunks and NFTs
Github: Github CryptoPunks EDA
Presentation Slides: CryptoPunks EDA Slides
About Author
Tam Trinh
View all posts by Tam Trinh >Related Articles
Leave a Comment
No comments found.