The Impact Mechanism of Algorithmic Transparency on User Trust in Intelligent Recommendation Systems of Internet Platforms

Whether intelligent recommendation algorithms based on computational systems can influence user trust in internet platforms has become a key issue in today’s digital society. However, relevant research remains limited, particularly in terms of systematic empirical investigation into the underlying mechanisms. Drawing on Fairness Heuristic Theory, this study introduces algorithmic fairness perception as a mediating variable and algorithmic literacy as a moderating variable to construct a conceptual model linking algorithmic transparency and user trust. A scenario-based experimental survey was conducted to systematically investigate the psychological mechanism through which algorithmic transparency affects user trust. The results demonstrate that overall, algorithmic transparency significantly enhances users’ trust in both the platform and its recommendation system. Furthermore, algorithmic fairness perception mediates this relationship, while algorithmic literacy positively moderates the effect of algorithmic transparency on fairness perception. These findings not only provide empirical support for understanding the relationship between algorithmic transparency and user trust but also offer practical insights for platform governance. Specifically, designers of recommendation algorithms should establish more standardized and effective transparency mechanisms and promote users’ algorithmic literacy to enhance their understanding of algorithmic logic, thereby strengthening trust at the cognitive level.