Improve Chit-Chat and QA Sentence Classification in User Messages of Dialogue System using Dialogue Act Embedding

1061

Authors

nan

Abstract

AbstractIn recent years, dialogue system is booming and widely used in customer service system, and has achieved good results. Viewing the conversation records between users and real customer service, we can see that the user’s sentences are mixed with questions about products and services, and chat with customer service. According to the experience of professionals, it is helpful in improving the user experience to mix some chats in customer service conversations. However, users’ questions are expected to be answered, while chatting is expected to interact with customer service. In order to produce an appropriate response, the dialogue system must be able to distinguish these two intentions effectively. Dialog act is a classification that linguists define according to its function. We think this information will help distinguishing questioning sentences and chatting sentences. In this paper, we combine a published COVID-19 QA dataset and a COVID-19-topic chat dataset to form our experimental data. Based on the BERT (Bidirectional Encoder Representation from Transformers) model, we build a question-chat classifier model. The experimental results show that the accuracy of the configuration with dialog act embedding is 16% higher than that with only original statement embedding. In addition, it is found that conversation behavior types such as “Statement-non-opinion”, “Signal-non-understanding” and “Appreciation” are more related to question sentences, while “Wh-Question”, “Yes-No-Question” and “Rhetorical-Question” questions are more related to chat sentences.

Keywords

nan

Full Content

�� (Dialog Act) �� Covid-19��Covid-19� �� BERT (Bidirectional Encoder Representation from Transformers) �� —�� (Dialog Act Embedding) �� 16%�� Statement-non-opinion�Signal-non- understanding�Appreciation �� Wh-Question�Yes-No- Question�Rhetorical-Question �� Abstract In recent years, dialogue system is booming and widely used in customer service system, and has achieved good results. Viewing the conversation records between users and real customer service, we can see that the user's sentences are mixed with questions about products and services, and chat with customer service. According to the experience of professionals, it is helpful in improving the user experience to mix some chats in customer service conversations. However, users' questions are expected to be answered, while chatting is expected to interact with customer service. In order to produce an appropriate response, the dialogue system must be able to distinguish these two intentions effectively. Dialog act is a classification that linguists define according to its function. We think this information will help distinguishing questioning sentences and chatting sentences. In this paper, we combine a published COVID-19 QA dataset and a COVID-19-topic chat dataset to form our experimental data. Based on the BERT (Bidirectional Encoder Representation from Transformers) model, we build a question-chat classifier model. The experimental results show that the accuracy of the configuration with dialog act embedding is 16% higher than that with only original statement embedding. In addition, it is found that conversation behavior types such as "Statement-non- opinion", "Signal-non-understanding" and "Appreciation" are more related to question sentences, while "Wh-Question", "Yes-No-Question" and "Rhetorical-Question" questions are more related to chat sentences. �� Keywords : Dialog act classification, Dialog system 1 �� (Task- Oriented)�� (Chit-Chat) �� Improve Chit-Chat and QA Sentence Classification in User Messages of Dialogue System using Dialogue Act Embedding Chi-Hsiang Chao ChingShin Academy Taipei, Taiwan 10835028@st.chjhs.tp.edu.tw Xi-Jie Hou ChingShin Academy Taipei, Taiwan 10935020@st.chjhs.tp.edu.tw Yu-Ching Chiu Department of Computer Science and Information Engineering National Central University Taoyuan, Taiwan crystalchiu@g.ncu.edu.tw The 33rd Conference on Computational Linguistics and Speech Processing (ROCLING 2021) Taoyuan, Taiwan, October 15-16, 2021. The Association for Computational Linguistics and Chinese Language Processing 138 �� (Question Answering) �� Google �� Siri �� (Chit-Chat) �� Google � Meena (Adiwardana, et al., 2020)�� (Question Answering) �� IBM � Watson (Gliozzo, et al., 2013)� �� (Dialogue Act) (Perkoff, E Margaret, 2021)�� 1 �� (Chit-Chat)�� (QA �� Question)�� Chit- Chat � QA �� Conventional-opening � Statement-non- opinion�� Chit-Chat�� Wh-Question � �� QA �� 2 � �� 2.1 Hybrid Dialogue System �� (Hosseini-Asl, et al., 2020) (Adiwardana, et al., 2020)�� (Moirangthem, …

Content truncated to 500 words

Labeled Text

Publication Details

Paper ID:: 1061
Year:: 2021
Publication:: TIPSTER
Publication Type:: conf
DOI:: nan

Links

Paper URL PDF Publication URL