The goal of this randomized controlled intervention is to develop and evaluate the effectiveness of the nurse-led with AI chatbot Individualized Symptom Management Education and Home-based Exercise (AI-ISME\&HE) based on The Theory of Symptom Self-Management (TSSM) on frailty, self-efficacy in cancer care, trAE, and quality of life in advanced gastric cancer patients receiving immunotherapy. The main questions it aims to answer are:
* Can AI-ISME\&HE reduces frailty in advanced gastric cancer patients receiving immunotherapy?
* Can AI-ISME\&HE increases quality of life in advanced gastric cancer patients receiving immunotherapy?
* Can AI-ISME\&HE increases self-efficacy in advanced gastric cancer patients receiving immunotherapy?
* Can AI-ISME\&HE reduces all-caused death in advanced gastric cancer patients receiving immunotherapy within two years? Researchers will compare usual routine care to see if AI-ISME\&HE can reduce frailty, all-caused death and increase quality of life and self-efficacy.
Participants will not be blinded, but the outcome evaluator will be blinded. The participants will be asked to complete three tasks for 12 weeks:
* Self-reporting treatment-related adverse events
* Interacting with AI-ISME\&HE for seeking information related to self-management educational materials, including symptom management with the most reported trAEs and prevention from frailty
* ViviFrail-based exercise