A Family-History Based Conversational AI System Using LLM

The proportion of single residents is steadily increasing, and the lives of single residents can experience relational disconnection and isolation, as well as limitations in social solidarity and integration. This study aimed to develop an AI conversation system that allows users to feel intimacy with the aim of improving the loneliness of single residents. Therefore, the system persona was set to family, and the system was created using the LLM model. The AI conversation system was developed at the pilot level, and user testing was conducted. Suppose the system is refined in the future and upgraded by reflecting user test opinions. In that case, it is believed that it will be possible to develop a system that allows users to feel closeness and, as a result, improves the loneliness and depression of single residents.