This study investigates research literature in the field of English language teaching (ELT) related to chatbots as an interactive user interface of the computer. It also discusses how to use chatbot builders to make ELT chatbots. Research on ELT chatbots in Korea has recently begun. At first, chatbot research focused on how to use general-purpose chatbots. Then, many studies changed their foci to how to make chatbots that were specialized in ELT. The analysis of the research literature shows that certain chatbot builders, such as Google's Dialogflow, have been primarily used, and ELT experts spend a considerable amount of time making chatbots. The way chatbots are built can be based on four models: pattern-matching model, dialog database model, knowledge base model, and language generation model. Based on this classification, domestic and foreign chatbot builders were analyzed. This study then suggests general-purpose chatbots for small talk or information retrieval activities, and customized chatbots built with chatbot builders for specific dialogs or tasks. The significance of this study can be found in that it proposes efficient ways to make chatbots in ELT.
- 주제 : 어문학분야 > 영어학
- 발행기관 : 한국멀티미디어언어교육학회
- 간행물 : Multimedia-Assisted Language Learning, 23권 4호