Research on Generative Dialogue Algorithm Based on Knowledge Graph

With the vigorous application of artificial intelligence technology in power grid business, electric power intelligent customer service robots have developed rapidly. However, in the dialogue process of intelligent customer service robots, there are often problems such as inadequate understanding of customer inquiries and difficulty in accurately capturing their true intentions, leading to the inability of intelligent customer service to fully realize their potential. In response to this issue, I have organized data related to power customer service, introduced knowledge graph technology, and considered factors such as customer service dialogue background and dialogue scenarios. I have studied a generative dialogue model based on knowledge graph to make the dialogue more targeted. At the same time, I have introduced background information methods to make the model better understand users and generate responses that are more in line with their needs, providing better services for users.