Predictive analysis of non fungible token price using deep learning

Abstract

A form of digital asset called Non-fungible tokens can represent a wide range of objects, such as pieces of art, collectibles, and in-game items. Non-fungible tokens are also commonly referred to by their acronym, NFTs. They are often kept within smart contracts that are hosted on a blockchain and are traded over the internet, where cryptocurrency is frequently used. The year 2021 has seen a meteoric rise in the acceptance of NFTs, which has resulted in exceptional sales in the market. Despite this, we still have little grasp of the overall structure of this market and how it evolved over time. Within the scope of this investigation, we investigate a dataset that contains 6.1 million transactions that involve 4.7 million non-fungible tokens and runs from 23rd June 2017 to 27th April 2021. The Ethereum and WAX blockchains are the primary sources of this information. Our analysis aims to achieve several objectives. In the first step of this process, we look into the statistical characteristics of the NFT market. In the second step of our process, we build a network that illustrates the relationships between different traders.We have noticed that traders frequently specialize in NFTs that are related with comparable objects, and they typically establish cohesive clusters with other traders who are involved in the trading of similar objects.Thirdly, we use clustering algorithms to organize the items that are associated with NFTs according to the visual qualities that distinguish them from one another. Our findings demonstrate that collections tend to consist of visually consistent objects.Finally, we investigate whether or not NFT sales may be forecasted by utilizing certain fundamental machine learning methods. According to the findings of our investigation, an NFT’s sales history and, to a lesser extent, its aesthetic characteristics can each serve as credible predictors of the price of that NFT. We are confident that these realizations will act as a driving force behind additional research on the creation, adoption, and trading of NFTs in a variety of different settings.