Developing Deep Learning-Based Chatbot for Health Dialogue

This study aims to build a prototype chatbot for consultation or dialogue about health. Online health consultations or dialogues are important in today's digital information era. The main problem of online health dialogue is the limited number of human resources in answering questions, where the number of people who ask will be far more than the number of people who ask. The availability of health workers who have the ability to answer questions is still very limited. One way to overcome this is to apply chatbot technology. A chatbot is an artificial intelligence application that can imitate a conversation or dialogue with a user in natural language. The chatbot allows communication via text or audio. In this study, an Indonesian-language health dialogue chatbot will be developed using the Transformer deep learning method. The results of the research will be developed into a prototype of the Health Dialog Chatbot Application that is able to assist users in understanding the health problems they face by having conversations with machines. It is hoped that the machine can interact naturally with the user, thus creating a comfortable feeling such as having a conversation or dialogue with a friend who is a health expert.