Abstract
Steemit is a blockchain-based online knowledge community that redefines incentive structures by rewarding participation with blockchain tokens. The platform uniquely balances content creation and consumption, but faces challenges in managing the vast influx of content, leading to potential information overload. Efficient recommender systems are crucial for ensuring content accessibility and preserving the recognition of individual contributions. In the specific context of Steemit, the implementation of these systems introduces complexities, particularly influencing token reward distribution and potentially affecting reward fairness. The research focuses on examining how recommender systems impact user engagement, content quality, and the fairness of rewards using simulation-based analysis with real-world Steemit data. It uncovers that while these systems can enhance the equity of rewards, they might compromise content quality. The findings underscore the importance of developing adaptable recommendation strategies that can sustain content excellence and equitable reward distribution, providing valuable insights for refining recommendation algorithms in blockchain-based knowledge communities.
Recommended Citation
Xu, Xiaoying; Li, Zhihong; Zhong, Xiaobin; and Zhang, Jie, "Simulation-Based Impact Analysis of Introducing Recommender Systems in Blockchain Token-Incentivized Knowledge Communities" (2026). ICEB 2026 Proceedings (Beijing, China). 82.
https://aisel.aisnet.org/iceb2026/82