Analysis of non-fungible token pricing factors with machine learning

Ho, Kin Ho, Hou, Yun, Chan, Tse Tin & Pan, Haoyuan (2022). Analysis of non-fungible token pricing factors with machine learning. In IEEE (Ed.). 2022 IEEE international conference on systems, man, and cybernetics (SMC). 2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Prague, Czech Republic. IEEE.

DOI

10.1109/SMC53654.2022.9945566

URI

http://hdl.handle.net/20.500.11861/9061

Type

Conference Paper

Abstract

Rarity is known to be a factor in the price of non-fungible tokens (NFTs). Most investors make their purchasing decisions based on the rarity score or rarity rank of NFTs. However, not all rare NFTs are associated with a higher price, especially for play-to-earn gaming NFTs. In this paper, we studied the top-ranked play-to-earn gaming NFTs on Axie Infinity. We found that, in addition to rarity, utility is also a significant factor influencing the price. Furthermore, we use utility as a predictor to predict the price of Axies using the XGBoost regressor. Our results reveal that, compared to using rarity-based predictors only, leveraging utility-based predictors can improve the prediction accuracy, thus highlighting utility as a price determinant for play-to-earn gaming NFTs.

Options

Analysis of non-fungible token pricing factors with machine learning

Availability at HKSYU Library