In this study, a social media chatbot was used to provide a continuous and real-time multimedia health education program on physical activity for the older adults, as well as supportive messages and feedbacks, to improve physical activity and exercise self-efficacy among the older adults. This study was a cluster randomized trial, and participants were recruited from community care stations and activity centers in Taipei City. The experimental group was involved in an 8-week, 5-day-a-week intervention with a total of 40 multimedia physical activity education programs, and users were provided with real-time feedback interactions and regular physical activity education guidelines, and self-administered questionnaires were used for data collection. The research instruments include basic personal information, International Physical Activity Questionnaire, Exercise Self-Efficacy Scale, Behavioral regulation in exercise questionnaire-2, Exercise Benefits/Barriers Scale, and the statistical methods will be descriptive statistics, independent sample t-test, paired sample t-test, and one-way analysis of covariance.

Inclusion Criteria:

1. Persons aged 60 to 85 years old living in Taipei City, healthy or with chronic conditions
2. Those who can listen, read and write Chinese
3. Those who have used smartphones in the past month and can use Line social software
4. Those who voluntarily participate in this study, can understand the research content after explanation and explanation, and sign the subject consent form

Exclusion Criteria:

1. Those who are advised by a doctor not to exercise
2. Those who must exercise under the supervision of medical personnel
3. Patients with severe cardiovascular disease
4. Anyone with disability or cognitive impairment
5. Persons living in long-term care institutions