Research on Interaction Design of Personalized Intelligent Exercise System for Seniors in China

Nowadays, the adherence of Chinese seniors with mobile device-based exercise interventions is poor. It is particularly important to explore the design factors that promote exercise behavior. This research innovatively integrated human-oriented design concept with third-generation artificial intelligence technology, and built a four-layer personalized intelligent exercise system (PIES) through the dual perspectives of focus groups (18 seniors) and in-depth interviews with experts (7 rehabilitation therapists). The system adopted federal learning framework to realize multi-source data fusion under privacy protection, integrated prescription generation engine of Transformer architecture and multi-modal emotion adaptation mechanism and to develop an interactive interface including four core modules, start training, assessment and preference, exercise prescription, and health status. The system greatly improved the exercise prescription compliance and reduced the abandonment rate in the first week. The research results provided a theoretical basis and technical framework for the establishment of seniors user characteristics database (120,000 samples have been collected) and the optimization of adaptive exercise prescription algorithm, which has important application value in the field of healthy aging.