This project investigates the intersection of Decentralized Finance (DeFi) systems, powered by distributed ledger technology (DLT), and the influence of social media influencers on digital asset markets, aiming to address the risks of market manipulation and insider trading.
The research has three main objectives. First, it seeks to identify the motivations and strategies of social media influencers, distinguishing organic market trends from manipulative activities. A mixed-methods approach, combining qualitative interviews and quantitative analysis, will uncover patterns in influencer behavior and their impact on market dynamics. Second, the project focuses on developing advanced machine learning algorithms. These will utilize techniques from network science, natural language processing, and on-chain data analysis to detect unique manipulation tactics in influencer-driven activities. Third, it aims to build predictive models that integrate social media and market data to forecast price shifts driven by influencer actions. These tools will help market participants and regulators anticipate and mitigate potential risks, bolstering market resilience.
The project introduces innovative methodologies, including graph-based network analysis and large language models (LLMs), to detect coordinated manipulative activities and track influencer narratives in real time. By bridging gaps in understanding social media’s impact on DeFi, the research will aid in mitigating manipulative behaviors, thereby supporting the stability and transparency of digital asset ecosystems. The findings are expected to benefit both regulators and market participants.