DOI QR코드

DOI QR Code

Natural Language Processing Model for Data Visualization Interaction in Chatbot Environment

챗봇 환경에서 데이터 시각화 인터랙션을 위한 자연어처리 모델

  • Received : 2020.09.21
  • Accepted : 2020.11.03
  • Published : 2020.11.30

Abstract

With the spread of smartphones, services that want to use personalized data are increasing. In particular, healthcare-related services deal with a variety of data, and data visualization techniques are used to effectively show this. As data visualization techniques are used, interactions in visualization are also naturally emphasized. In the PC environment, since the interaction for data visualization is performed with a mouse, various filtering for data is provided. On the other hand, in the case of interaction in a mobile environment, the screen size is small and it is difficult to recognize whether or not the interaction is possible, so that only limited visualization provided by the app can be provided through a button touch method. In order to overcome the limitation of interaction in such a mobile environment, we intend to enable data visualization interactions through conversations with chatbots so that users can check individual data through various visualizations. To do this, it is necessary to convert the user's query into a query and retrieve the result data through the converted query in the database that is storing data periodically. There are many studies currently being done to convert natural language into queries, but research on converting user queries into queries based on visualization has not been done yet. Therefore, in this paper, we will focus on query generation in a situation where a data visualization technique has been determined in advance. Supported interactions are filtering on task x-axis values and comparison between two groups. The test scenario utilized data on the number of steps, and filtering for the x-axis period was shown as a bar graph, and a comparison between the two groups was shown as a line graph. In order to develop a natural language processing model that can receive requested information through visualization, about 15,800 training data were collected through a survey of 1,000 people. As a result of algorithm development and performance evaluation, about 89% accuracy in classification model and 99% accuracy in query generation model was obtained.

스마트폰의 보급으로 인해 개인화된 데이터를 활용하고자 하는 서비스들이 증가하고 있다. 특히, 헬스케어와 관련된 서비스들은 다양한 데이터를 다루며, 이를 효과적으로 보여주기 위해 데이터 시각화 기법을 활용하고 있다. 데이터 시각화 기법이 활용되면서 자연스럽게 시각화에서의 인터랙션 또한 함께 강조되고 있다. PC 환경에서 데이터 시각화에 대한 인터랙션은 마우스로 이루어지기 때문에, 데이터에 대한 필터링이 다양하게 제공되고 있다. 반면, 모바일 환경에서의 인터랙션은 화면의 크기가 작고, 인터랙션 가능 여부를 인지하기 어려워 버튼 터치 방식으로 앱에서 제공하는 제한된 시각화만을 제공받을 수 있다. 이러한 모바일 환경에서의 인터랙션 한계를 극복하기 위해, 챗봇과의 대화를 통해 데이터 시각화 인터랙션을 가능하게 하여 사용자들에게 개개인의 데이터를 다양한 시각화를 통해 확인할 수 있도록 하고자 한다. 이를 위해서는 사용자의 질의를 쿼리로 변환하여, 주기적으로 데이터를 축적하고 있는 데이터베이스에서 변환된 쿼리를 통해 결과 데이터를 불러올 수 있어야 한다. 자연어를 쿼리로 변환하는 연구는 현재 많이 이루어지고 있지만, 시각화를 기반으로 하여 사용자의 질의를 쿼리로 변환하는 연구에 대해서는 아직 이루어지지 않았다. 따라서, 본 논문에서는 사전에 데이터 시각화 기법이 정해진 상황에서의 쿼리 생성에 초점을 맞추고자 한다. 지원하는 인터랙션은 태스크 x-축 값에 대한 필터링 및 두 그룹 간 비교이다. 테스트 시나리오는 걸음 수에 대한 데이터를 활용하였으며, x-축 기간에 대한 필터링은 바 그래프, 두 그룹간 비교는 라인 그래프로 나타내었다. 시각화를 통해 요청한 정보를 제공받을 수 있는 자연어처리 모델을 개발하기 위해 1,000명을 대상으로 한 설문조사를 통해 약 15,800개의 학습 데이터를 수집하였다. 알고리즘 개발 및 성능 평가를 진행한 결과, 분류 모델에서는 약 89%, 쿼리 생성 모델에서는 약 99% 정확도를 보였다.

Keywords

References

  1. E. S. Kim, "Affecting factors of mobile health care service usage and efficient utilization plan," Korean Journal of Health Education and Promotion, Vol.34, No.2, pp.41-52, 2017. https://doi.org/10.14367/kjhep.2017.34.2.41
  2. L. S. Lee, S. H. Lee, J. S. Jeong, and K. Y. Noh, "Psychological factors influencing continuous use of mobile healthcare applications," Journal of Digital Convergence, Vol.15, no.7, pp.445-456, 2017. https://doi.org/10.14400/JDC.2017.15.7.445
  3. J. H. Park, "Smart Health Care Prospective Market Trends and Strategies" [Internet], https://news.kotra.or.kr/user/reports/kotranews/20/usrReportsView.do?reportsIdx=10985
  4. Y. H. Park and J. Y. Yun, "Design guidelines for data visualization of smart band: Focused on fitbit," The Korean Society of Science & Art, Vol.30, pp.141-149, 2017. https://doi.org/10.17548/ksaf.2017.09.30.141
  5. E. J. No, "Visualization study of healthcare data: Focusing on mobile healthcare services," M.A. Degree Thesis, University of Ewha Womans at Seoul, Korea, 2015.
  6. Liu, Zhicheng and J. Stasko, "Mental models, visual reasoning and interaction in information visualization: A top-down perspective," in IEEE Transactions on Visualization and Computer Graphics, Vol.16, No.6, pp.999-1008, 2010. https://doi.org/10.1109/TVCG.2010.177
  7. K. Blumenstein, C. Niederer, M. Wagner, G. Schmiedl, A. Rind, and W. Aigner, "Evaluating information visualization on mobile devices: Gaps and challenges in the empirical evaluation design space," in Proceedings of the Sixth Workshop on Beyond Time and Errors on Novel Evaluation Methods for Visualization, 2016.
  8. J. Y. Kim, "Analysis and design of mobile chatbot interface," M.Eng Thesis, University of Seoul National at Seoul, Korea, 2017.
  9. B. J. Kim. "Cognitive and emotional UX design methodology for mobile chatbot," Korea Science & Art Forum, Vol.34, 2018.
  10. D. Y. Chang and C. K. Lee, "A study of use intention of chatbot using the extended theory of planned behavior:Focusing on the role of interaction," Journal of Tourism and Leisure Research, Vol.31, No.8, pp.433-454, 2019.
  11. V. Zhong, C. Xiong, and R. Socher, "Seq2SQL: Generating structured queries from natural language using reinforcement learning," CoRR, abs/1709.00103, 2017.
  12. Y. Sun, D. Tang, N. Duan, J. Ji, G. Cao, X. Feng, B. Qin, T. Liu, and M. Zhou, "Semantic parsing with syntax- and table-aware SQL generation", in Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, pp.361-372, 2018.
  13. P. Wang, T. Shi, and C. K. Reddy, "Text-to-SQL Generation for Question Answering on Electronic Medical Records," in Proceedings of The Web Conference, pp.350-361, 2020.
  14. I. Ko and H. Chang, "Interactive Visualization of Healthcare Data Using Tableau," Healthcare Informatics Research, Vol.23, No.4, pp.349-354, 2017. https://doi.org/10.4258/hir.2017.23.4.349
  15. D. J. Janvrin, R. L. Raschke, and W. N. Dilla, "Making sense of complex data using interactive data visualization," Journal of Accounting Education, Vol.32, No.4, pp.31-48, 2014. https://doi.org/10.1016/j.jaccedu.2014.09.003
  16. J. S. Yi, Y. A. Kang, J. Stasko, and J. A. Jacko, "Toward a deeper understanding of the role of interaction in information visualization," IEEE Transactions on Visualization and Computer Graphics, Vol.13, No.6, pp.1224-1231, Nov.-Dec. 2007. https://doi.org/10.1109/TVCG.2007.70515
  17. DeepLearning.AI, Improving Deep Neural Networks: Hyperparameter tuning, Regularization and Optimization [Internet], https://ko.coursera.org/lecture/deep-neural-network/rmsprop-BhJlm
  18. D. P. Kingma, and J. Ba, "Adam: A Method for Stochastic Optimization," in arXiv e-prints, 2014.
  19. M. J. Kim, Adam Optimizer [Internet], http://mjgim.me/2018/01/22/adam.html