Personalized Dialogue Generation Method of Chat Robot Based on Topic Perception

• Published in: Multimedia Technology and Enhanced Learning Vol. 446; pp. 549 - 560
• Main Author: Li, Junmei
• Format: Book Chapter
• Language: English
• Published: Switzerland Springer 2022; Springer Nature Switzerland
• Series: Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
• Subjects: Chatbot; Dialogue generation; Personalization
• ISBN: 3031181220; 9783031181221
• ISSN: 1867-8211; 1867-822X
• DOI: 10.1007/978-3-031-18123-8_43

Human-Computer interaction system is a significant research direction in the field of human-computer interaction, and the research of open domain chat robot has received extensive attention. There are many problems in the existing chat robot: lack of personalized features, resulting in the process of the same chat, and the conversation has nothing to do with the topic. Therefore, a method of creating personalized conversation based on topic perception is proposed, and a personalized conversation model based on topic perception is designed. Semantic analysis and text similarity calculation are needed to build a conversation model. Based on the dialogue model, the training robot collects the corpus data related to the subject, convolves the corpus data related to the subject, and carries out the topic perception training. Finally, a personalized dialogue mechanism is established to generate personalized dialogue. Through experimental comparison, it is proved that the dialogue generated by this method is more suitable for the chat topic.