Predicting NFT Classification with GNN: A Recommender System for Web3 Assets

The development of effective recommender systems for Web3 assets, such as the Non-Fungible Token (NFT), requires concentration along with the growth of popularity and heterogeneity in many potential applications such as Web3 gaming and NFT rental markets, the requirements of predicting rNFT classification desire a practical solution. In this paper, we make use of the referable NFT (rNFT¹¹In this work, rNFT mainly refers to the EIP-5521 protocol and corresponding formed network/topology [1], while NFT is used in the context of a single node, node sets, or products that align with the EIP-5521 protocol.) standard [2], indexed EIP-5521, to construct an rNFT classification framework leveraging Graph Neural Network (GNN), an emerging branch of Deep Learning (DL), which learns on the inherent topology of graph-based data. In particular, we first transform the rNFT backward and onward reference relationship to a Direct Acyclic Graph (DAG) and model appropriate node and edge features from rNFT metadata and associated token transactions. Next, a multi-layer GraphSage model is designed to include the collected features for the learning process. In this way, the model takes into account graph topology together with features to classify both the existing and incoming NFT nodes in a supervised way. We also give comprehensive elaboration on the architecture of the new GNN-based recommender system with discussions in regard to its characteristics and challenges. Furthermore, we expect to conduct extensive experiments, by presenting an initial plan, to show the feasibility and efficacy of our system.

I. Introduction

Non Fungible Tokens (NFTs) [3] is an Ethereum token standard (ERC-721 [4]) that focus on defining unique digital assets. NFTs are aligned with the natural principle of Ethereum ecosystems [5] in which trading an NFT is like making normal transactions on-chain. It specifies each token with an identifier by encoding a string of tokenID within smart contracts. This brings uniqueness to each piece of attached data within the corresponding transactions. Not surprisingly, NFT hits the markets with intensive attention as it has impacted a wide variety of areas such as collectibles, art, games, and intellectual properties. To date²

Data captured from https://nonfungible.com/reports [Nov 2022].

(Q1&Q2&Q3 2022), the total traded volume reaches up to 20,479,676,915 USD and the volume of sales is 33,651,380. Involved buyers and sellers are separately 3,647,060 and 2,118,542. Active wallets also boom up to 4,285,553. NFTs, as proved by data, have formed a new place that brings traditional markets into new Web3 spaces [6].