Chatbot solutions for health care support in Vietnamese region
Keywords:
Chatbot NLP, Healthcare, Intent categorization, Information extraction, Machine learningAbstract
Nowadays, Chatbot has been being applied a lot in the e-commerce field. For industries such as healthcare, Chatbot has not yet been widely applied because of its high expertise and need of expert support. Chatbots supporting Vietnamese nowadays often apply two traditional models, according to Menu/Button scenario and keyword identification, which exist many limitations on efficiency and ability to interact with users. This article builds a Chatbot solution to support health care based on the DIET model to classify intent and select information to analyze users' words in Vietnamese. The main intentions classified in the experiment include: asking about symptoms, asking about the cause, asking how to prevent, and asking for treatment. The selected information includes: name, sex, disease. The efficiency of the model when conducting experiments reached more than 95% accuracy on the test data set.
References
Nadarzynski, Tom et al. “Acceptability of artificial intelligence (AI)-led chatbot services in healthcare: A mixed-methods study.” Digital health, Vol. 5 (2019).
V. M. Do, D. N. Tran, V. A. Nguyen, T. H. V. Ha, “Xây dựng giải pháp Chatbot bán hàng tự động iBotsale trên miền Tiếng Việt dựa trên trích trọn thông tin và phân loại ý định”, Hội thảo: Một số vấn đề chọn lọc của Công nghệ thông tin và truyền thông (2020).
Delhi, “Review of chatbot design and trends”, Artificial Intelligence and Speech Technology (2019).
J. Devlin, M. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding”, In Proceedings of the 2019 Conference of the NACL, Vol 1 (2019), pp. 4171–4186.
M. E. Peters et al, “Deep contextualized wordrepresentations”, arXiv:1802.05365 [cs.CL] (2018).
G. Lample and A. Conneau, “Cross-lingual language modelpretraining”, CoRR, (2019).
M. E. Peters et al, “Dissecting contextual word embeddings: Architectureand representation”, CoRR, abs/1808.08949, (2018).
Z. Yang et al, “Xlnet: Generalized autore-gressive pretraining for language understanding”, CoRR, abs/1906.08237 (2019).
T. Bunk, D. Varshneya, V. Vlasov, A. Nichol, “DIET: Lightweight Language Understanding for Dialogue Systems”, arXiv: 2004.09936 [cs.CL] (2020).
N. Elvina, A. Rusli, S. Hansun, “Line Bot Chat Filtering using Naïve Bayes Algorithm”, IJITEE, Vol. 8 (2019), pp. 2278-3075.
A. Nigam, P. Sahare, K. Pandya, “Intent Detection and Slots Prompt in a Closed-Domain Chatbot”, arXiv:1812.10628 [cs.CL] (2019).
Vu Anh, “Underthesea - Vietnamese NLP Toolkit” (2019).
Q. T. Hoang, C. P. K. Phung, N. T. Bui, T. P. D. Chu, & D. T. Tran, “Cow behavior monitoring using a multidimensional acceleration sensor and multiclass SVM”, International Journal of Machine Learning and Networked Collaborative Engineering (2018), pp. 110-118.
D. T. Tran, & V. T. Nguyen, “Reliable fall detection system using a 3-DOF accelerometer and cascade posture recognitions”, In Signal and Information Processing Association Annual Summit and Conference (APSIPA) (2014), pp. 1-6.
C. P. K. Phung, T. L. Ton, D. C. Nguyen, & D. T. Tran, “Performance evaluation of a multi-stage classification for cow behavior”, In 2018 2nd International Conference on Recent Advances in Signal Processing, Telecommunications & Computing (2018), pp. 121-125.
D. N. Tran, T. N. Nguyen, C. P. K. Phung, & D. T. Tran, “An iot-based design using accelerometers in animal behavior recognition systems”, IEEE Sensors Journal (2021).
