<p>Group recommender systems aim to provide recommendations that satisfy the collective interests of multiple users with potentially conflicting preferences. While prior research has explored aggregation techniques and influence-aware models, few have effectively leveraged implicit, temporal influence derived from user interaction sequences to balance fairness, accuracy, and diversity in group recommendations. In this paper, we propose a novel influence-based aggregation framework that dynamically constructs a directed influence graph based on the order of user ratings within each group. By computing user influence scores and integrating them into the recommendation process, our approach prioritizes the preferences of trustworthy users without requiring external influence data. A key innovation of the proposed method is the introduction of a tunable parameter β, which enables adaptive control over the trade-off between diversity, fairness, and accuracy. Specifically, increasing β enhances diversity, whereas lowering it improves accuracy and fairness, providing a flexible mechanism to tailor recommendations to different application needs. Extensive experiments across multiple datasets demonstrate that our approach significantly outperforms baseline methods in simultaneously optimizing fairness, diversity, and accuracy, offering a practical and interpretable solution for real-world group recommendation scenarios.</p>