NFT Price Analysis Data Challenge
Are you ready to forecast NFT prices & win? We’re announcing our NFT Price Analysis data challenge due this April 30th, 2023 at 11:59 PM UTC! Contestants are asked to submit data analytics reports and machine learning models to analyze the floor prices of NFTs.
Accurate predictions of NFT prices are crucial for investment decisions and estimating risk for NFT financial products. In this challenge, participants not only estimate NFT floor prices, but also showcase their AI skills. Contestants must perform an exploratory analysis on provided or participant-found data and building a machine-learning model that predicts the floor price of an NFT.
Ocean Protocol’s NFT Price Analysis Prizes
The prize pool for this competition is $5,000 USD payable in OCEAN, with the top three participants receiving cash prizes as follows:
1st place: $2,500
2nd place: $1,500
3rd place: $1,000
NFT Price Analytics Hackathon
All the requirements for this challenge can be found on the NFT Floor Price Prediction Challenge page. Participants can use any data they wish — static data or streams, free or priced, raw data or feature vectors, published on Ocean Market or not. There are also links to some data feeds and AI modeling approaches that may be helpful in the challenge’s description.
We are providing a dataset from Transpose data that shows all of the transactions for the NFT collections Bored Ape Yacht Club, Mutant Ape Yacht Club, Azuki, Moonbirds, and Otherdeed. Participants should pick two NFT collections to analyze.
The submission deadline for the reporting part of the competition is April 30th, 2023, at 23:59 UTC. Participants must submit their reports by the deadline to be eligible.
Evaluation Criteria
The winner of the data report competition will be judged based on the following criteria:
Data Analysis (50 points)
Select five questions below to answer as part of your analysis. Your answer should include quantitative and qualitative elements.
Choose 2 NFT collections to evaluate
- Analyze how the number of daily transactions for the collections has changed over time.
- Provide a visual overview of the NFT collections of your choice and its characteristics (e.g. size, type of NFTs, date range)?
- Can you identify any clusters or groups of NFTs within the collection based on their attributes or characteristics? If so, what are the key features of these groups?
- How does the price of NFTs with similar levels of rarity compare across different NFT collections?
- Can you identify any correlations between the characteristics of the NFTs within the collection and their prices? If so, what are the most important factors influencing the prices?
- Have any traits increased in value over time, and if so, which ones?
- Which NFTs in a collection were most profitable when they were flipped (bought and sold quickly)
- Determine the correlation between the number of transactions in a collection and its floor price
- Determine the correlation between the number of transactions in a collection and the price of ETH
- What are the most liquid traits (those with the most sales) for each collection?
Report (25 points):
Submit a report describing the above findings. Make sure to include qualitative insights in addition to quantitative ones. Reports will be evaluated on presentation structure, approach, content, and completeness.
Predict Model (25 points)
Use your findings above to develop a machine learning model that can be used to determine the current floor price of the rarest NFT in your chosen collection. Based on your model, is the NFT underpriced or overpriced? You will be judged on your choice of features and model.
Bonus (10 points)
Source your own dataset to create an additional lens to analyze the NFT collections. This could include datasets about the top wallet holders, Twitter sentiment, etc.
Description of the Transpose NFT Dataset Provided:
- indexer_id: Unique identifier for the NFT sale or collection.
- __confirmed: Flag indicating whether the transfer has been confirmed.
- __block_number: The block number at which the NFT sale occurred.
- block_number: The block number at which the NFT sale occurred.
- log_index: The log index at which the NFT sale occurred.
- transaction_hash: The transaction hash at which the NFT sale occurred.
- timestamp: The timestamp of the NFT sale (in ISO-8601 format).
- exchange_name: The name of the exchange that hosted the NFT sale.
- contract_version: The version of the exchange contract that hosted the NFT sale (e.g. wyvern or seaport for OpenSea).
- aggregator_name: The name of the aggregator used in the NFT sale (null if no aggregator was used).
- contract_address: The contract address of the NFT that was sold or the collection.
- token_id: The token ID of the NFT that was sold or the NFT.
- is_multi_token_sale: Whether the sale is a multi-token sale, including more than one unique NFT for the given payment.
- multi_token_sale_index: The index of the sale within the multi-token sale (will be 0 if not a multi-token sale).
- price: The total value of this sale in the payment token (in Wei).
- usd_price: The total value of this sale in USD.
- eth_price: The total value of this sale in ETH.
- native_price: The total value of this sale in the native token (ETH).
- payment_token_address: The address of the token used to pay for this sale (null if the native token, ETH).
- quantity: The quantity of tokens sold (will only be greater than 1 for ERC-1155 NFTs).
- seller_address: The address of the account that sold the NFT.
- buyer_address: The address of the account that bought the NFT.
- royalty_fee: The decimal-adjusted royalty fee paid to the creator of the NFT.
- platform_fee: The decimal-adjusted platform fee paid to the exchange that facilitated the NFT sale.
- minted_timestamp: The NFT’s mint timestamp (in ISO-8601 format) or the collection’s timestamp of creation.
- supply: The NFT’s supply (0 if NFT has been burned)
- name: The NFT’s name.
- description: The NFT’s description or the collection’s description.
- image_url: The NFT’s image URL or the collection’s image URL in the Transpose CDN.
- external_url: The NFT’s website URL or the collection’s website URL.
- media_url: The NFT’s additional media URL.
- properties: The NFT’s properties (also referred to as attributes or traits) or flattened properties of a collection.
- metadata_url: The NFT’s metadata URL.
- last_refreshed: The timestamp at which the collection or NFT was last refreshed by the Transpose backend (in ISO-8601 format).
- flattened_properties: Flattened properties of a collection.
- __updated_block_number: The block number at which the collection was last updated.
- collection.name: The name of the collection.
Conclusion
Join the community of data scientists and NFT enthusiasts, and take part in the NFT Floor Price Prediction Challenge today!
About Ocean Protocol
Ocean was founded to level the playing field for AI and data. Ocean tools enable people to privately & securely publish, exchange, and consume data.
Follow Ocean on Twitter or Telegram to keep up to date. Chat directly with the Ocean community on Discord. Or, track Ocean progress directly on GitHub.
