사용자와 실시간으로 감성적 소통이 가능한 한국어 챗봇 시스템 개발
Development of a Korean chatbot system that enables emotional communication with users in real time
1
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
In this study, the creation of emotional dialogue was investigated within the process of developing a robot's natural language understanding and emotional dialogue processing. Unlike an English-based dataset, which is the mainstay of natural language processing, the Korean-based dataset has several shortcomings. Therefore, in a situation where the Korean language base is insufficient, the Korean dataset should be dealt with in detail, and in particular, the unique characteristics of the language should be considered. Hence, the first step is to base this study on a specific Korean dataset consisting of conversations on emotional topics. Subsequently, a model was built that learns to extract the continuous dialogue features from a pre-trained language model to generate sentences while maintaining the context of the dialogue. To validate the model, a chatbot system was implemented and meaningful results were obtained by collecting the external subjects and conducting experiments. As a result, the proposed model was influenced by the dataset in which the conversation topic was consultation, to facilitate free and emotional communication with users as if they were consulting with a chatbot. The results were analyzed to identify and explain the advantages and disadvantages of the current model. Finally, as a necessary element to reach the aforementioned ultimate research goal, a discussion is presented on the areas for future studies.
Abstract
In this study, the creation of emotional dialogue was investigated within the process of developing a robot's natural language understanding and emotional dialogue processing. Unlike an English-based dataset, which is the mainstay of natural language processing, the Korean-based dataset has several shortcomings. Therefore, in a situation where the Korean language base is insufficient, the Korean dataset should be dealt with in detail, and in particular, the unique characteristics of the language should be considered. Hence, the first step is to base this study on a specific Korean dataset consisting of conversations on emotional topics. Subsequently, a model was built that learns to extract the continuous dialogue features from a pre-trained language model to generate sentences while maintaining the context of the dialogue. To validate the model, a chatbot system was implemented and meaningful results were obtained by collecting the external subjects and conducting experiments. As a result, the proposed model was influenced by the dataset in which the conversation topic was consultation, to facilitate free and emotional communication with users as if they were consulting with a chatbot. The results were analyzed to identify and explain the advantages and disadvantages of the current model. Finally, as a necessary element to reach the aforementioned ultimate research goal, a discussion is presented on the areas for future studies.
참고문헌 (23)
- L. C. De Silva, "Facial emotion recognition using multi-modal information", Proceedings of ICICS, 1997 Int. Conf. on Inf., Commun. Signal Process.. Theme: Trends Inf. Syst. Eng. Wirel. Multimed. Commun. (Cat., IEEE), 1, pp.397~401 (1997)
- S. Kim, "Realtime emotion detection system using speech: Multi-modal fusion of different timescale features", 2007 IEEE 9th Workshop Multimed. Signal Process, pp.48~51 (2007)
- A. Dufaux, "Automatic sound detection and recognition for noisy environment", 2000 10th Eur. Signal Process. Conf., pp.1~4 (2000), IEEE
- D. S. Moirangthem, "Chat Discrimination for Intelligent Conversational Agents with a Hybrid CNNLMTGRU Network", Proceedings of The 3rd Workshop Represent. Learn. NLP, pp.30~40 (2018), NLP
- D. Kang, "An intelligent chat bot agent system that can provide various services according to user’s intention", Int. Conf. Electronics, Inf. Commun. (ICEIC)2019, pp.1010~1011 (2019)
- T. Park, "Investigation of Embedding Methods for Korean Gag Chat-bot", The 3rd Int. Conf. Consum. Electronics(ICCE)Asia, pp.42~44 (2018)
- D. S. Moirangthem, "Improving Text Classification with Deep Recurrent Neural Networks", 18th China-Japan-Korea Joint Workshop Neurobiol. Neuroinformatics, pp.7(1)~7(1) (2018)
- G. C. Sergio, "Robust Sentiment Classification in Noisy User-generated Text with BERT-based model", Thirty-seventh Int. Conf. Mach. Learn.(ICML 2020), pp.24(1)~24(1) (2020)
- G. C. Sergio, "Stacked DeBERT: All attention in incomplete data for text classification", Neural Netw, 136, pp.87~96 (2021)
- G. C. Sergio, "Emotional video to audio transformation using deep recurrent neural networks and a neuro-fuzzy system", Math. Probl. Eng, 2020, pp.8478527(1)~8478527(15) (2020)
- D. Kang, "Seq-DNC-seq : Context Aware Dialog Generation System Through External Memory", Int. Joint Conf. Neural Netw(IJCNN), pp.1~8 (2019)
- M. A. Kiasari, "Joint moment-matching autoencoders", Neural Netw, 106, pp.185~193 (2018)
- D. S. Moirangthem, "Abstractive summarization of long texts by representing multiple compositionalities with temporal hierarchical pointer generator network", Neural Netw., 124, pp.1~11 (2020)
- D. S. Moirangthem, "Hierarchical and lateral multiple timescales gated recurrent units with pre-trained encoder for long text classification", Expert Syst. Appl, 165, pp.113898(1)~113898(8) (2021)
- R. Zandieand, "Emptransfo: A multi-head transformer architecture for creating empathetic dialog systems", The Thirty-Third Int.Flairs Conf. 2020, pp.276~281 (2020)
- T. Wolf, "Transfertransfo:A transfer learning approach for neural network based conversational agents", arXiv preprint, 1901.08149~1901.08149(6) (2019)
- R. C. Streijl, "Mean opinion score(MOS)revisited : methods and applications, limitations and alternatives", Multimedia Systems, 22(2), pp.213~227 (2016)
- K. Papineni, "Bleu: a method for automatic evaluation of machine translation", Proc. of 40th Annu. Meet. Association Comput. Linguist, pp.311~318 (2002)
- D. Adiwardana, "Towards a human-like open-domain chatbot", arXiv preprint, 2001.09977~2001.09977(38) (2020)
- K. Clark, "Electra:Pre-training text encoders as discriminators rather than generators", arXiv preprint, 2003.10555~2003.10555 (2020)
- E. Wallace, "Universal adversarial triggers for attacking and analyzing NLP", arXiv preprin, 1908.07125~1908.07125 (2019)