This paper presents a study on the development of a question-answering chatbot for groupware usage using BERT (Bidirectional Encoder Representations from Transformers). The primary objective of this research is to reduce the workload of customer suppo...
This paper presents a study on the development of a question-answering chatbot for groupware usage using BERT (Bidirectional Encoder Representations from Transformers). The primary objective of this research is to reduce the workload of customer support teams, provide 24/7 real-time assistance, and ensure consistent and accurate responses to user queries. Additionally, this study emphasizes the effectiveness of specialized models such as KLUE BERT for question-answering tasks related to groupware usage. Experimental results demonstrate that the KLUE BERT-based model exhibits high accuracy and consistent performance in Korean question-answering tasks, confirming its potential to significantly enhance the efficiency of customer support. keyword : Chatbot, BERT, Multilingual BERT, KLUE BERT, Natural Language Processing * A thesis for the degree of Master in February 2025