What is it about?
The human-machine dialogue system allows machines to interact with humans in natural language. If the machine can correctly use external knowledge during the dialogue with humans, it will generate more informative and interesting responses, thereby enhancing the practicality of the dialogue system. For machines, external knowledge can be a structured knowledge base or unstructured text. This survey introduces the research status of unstructured text-driven dialogue systems. Specifically, this article defines related concepts, introduces published data resources, defines two main research modes, classifies and summarizes the research methods/models/evaluations under these two modes, compares the experimental results of the models, and prospects for possible future research directions. Therefore, this article provides assistant to researchers of dialogue systems.
Photo by Possessed Photography on Unsplash
Why is it important?
It is the first systematic survey about Unstructured Text Enhanced Open-Domain Dialogue System.
Read the Original
This page is a summary of: Unstructured Text Enhanced Open-Domain Dialogue System: A Systematic Survey, ACM Transactions on Information Systems, January 2022, ACM (Association for Computing Machinery), DOI: 10.1145/3464377.
You can read the full text:
The following have contributed to this page