Leveraging blockchain and graph neural network to enhance carbon emission trading: A decentralized and trustworthy approach

Abstract

Carbon dioxide emissions are a major cause of climate change and its negative impacts on the environment and human well-being. To address this problem, a blockchain-based decentralized system for carbon emission trading is proposed, allowing anyone to trade carbon allowances as a commodity. The proposed system leverages the advantages of blockchain technology, such as transparency, immutability, and security, to enable efficient and trustworthy transactions among peers without intermediaries. Additionally, the system provides flexibility by enabling customization of permits, represented as Non-Fungible Tokens, thereby enhancing market accessibility and engagement. A graph neural network for identity inference is introduced to infer the attributes of accounts on the blockchain, such as their type and behavior. This can help detect and prevent illegal or malicious activities on the chain, as well as understand and improve user behavior patterns and preferences. The aim is to increase market inclusiveness and diversity, reduce transaction costs and carbon price volatility, and enhance the traceability and verifiability of carbon emissions.