Deep learning-driven personalized recommendations and layout optimization for UI interaction design

Deep learning-driven personalized recommendations and layout optimization for UI interaction design

연구 분야: Artificial Intelligence

학회: Progress in Artificial Intelligence

With the rapid growth of mobile Internet, UI design plays a crucial role in human-computer interaction, directly impacting user experience. Deep learning introduces new possibilities for personalized recommendations and layout optimization in UI design. This paper investigates deep learning-driven approaches to enhance mobile application UI by optimizing layout and recommendations through data-driven methods, ultimately improving user satisfaction and operational efficiency. First, the article outlines the data-driven UI design approach and analyzes the current applications in big data visualization and user behavior analysis. Through in-depth analysis of users’ historical behaviors, preferences, and interaction patterns, this article proposes a personalized recommendation system framework based on deep learning. The system is able to recommend the UI design and interaction layout that best meets the user’s needs in real time by collecting the user’s historical behavioral data and interaction features, combined with a deep neural network (DNN) model. Further, this paper optimizes the UI design by generative adversarial network (GAN) and proposes an innovative UI layout generation method, which can automatically generate UI layout styles with high user acceptance. The experimental results show that the deep learning-driven recommender system can not only effectively improve the personalization level of UI design, but also significantly improve the user interaction efficiency and satisfaction in multiple real-world application scenarios.