Scholarlink

To make mHealth apps more effective and engaging, personalization has emerged as a key approach. One promising method to achieve personalization is reinforcement learning (RL), a machine learning method well-suited for sequential decision-making. In this study, we introduce and apply an offline RL framework adapted to the novel setting of mHealth personalization. This framework includes algorithms that leverage historical user data and there is an emphasis on explainable AI to ensure interpretability. By using an offline approach, the framework enables the application of RL without the costs or risks associated with real-world experiments.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Personalizing mHealth Apps with Offline Reinforcement Learning: A Case Study on Mental Health App Adherence

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Rutger van der Linden,
  • Khadicha Amarti,
  • Marketa Ciharova,
  • Annet Kleiboer,
  • Heleen Riper,
  • Aneta Lisowska,
  • Mark Hoogendoorn

摘要

To make mHealth apps more effective and engaging, personalization has emerged as a key approach. One promising method to achieve personalization is reinforcement learning (RL), a machine learning method well-suited for sequential decision-making. In this study, we introduce and apply an offline RL framework adapted to the novel setting of mHealth personalization. This framework includes algorithms that leverage historical user data and there is an emphasis on explainable AI to ensure interpretability. By using an offline approach, the framework enables the application of RL without the costs or risks associated with real-world experiments.