딥러닝 모델 기반의 수두 발생 예측

다국어 초록 (Multilingual Abstract)

Chicken pox is a highly diffuse disease, and the need for surveillance research to predict it is increasing. Initially used CDC data takes at least a week to a month for this data to be confirmed. So there is a need to predict chicken pox using web data that can be collected in real time. Chicken pox, unlike other infectious diseases, appears frequently in web data regardless of actual outbreak data. Therefore, their linear relationship is not clear enough to be applied to existing linear regression models. In this paper, we predict chicken pox through deep learning model that can model nonlinear relationship. In addition, the prediction accuracy is improved by extracting the keyword related to the outbreak of chicken pox. Finally, the LSTM prediction model was able to predict the chicken pox for a longer period of time and had the highest correlation coefficient of 0.97114. The root mean square error was 341.01547, which was overwhelmingly smaller than the linear regression model.

목차 (Table of Contents)

• Abstract
• 1. 서론
• 2. 관련연구
• 3. 배경지식
• 4. 실험환경 및 데이터 수집
• 5. 키워드 추출
• 6. 사전연구
• 7. 딥러닝을 통한 수두 예측
• 7. 예측 모델 비교 및 평가
• 8. 결론
• References

참고문헌 (Reference)

1 J. Kim, "Weekly ILI patient ratio change prediction using news articles with support vector machine" 20 (20): 259-, 2019

2 A. Culotta, "Towards detecting influenza epidemics by analyzing Twitter messages" 115-122, 2010

3 R. Chunara, "Social and news media enable estimation of epidemiological patterns early in the 2010 Haitian cholera outbreak" 86 (86): 39-45, 2012

4 T. Chai, "Root mean square error(RMSE)or mean absolute error(MAE)?" 7 : 1525-1534, 2014

5 H. Achrekar, "Predicting flu trends using twitter data" 702-707, 2011

6 J. Benesty, "Noise Reduction in Speech Processing" Springer 1-4, 2009

7 F. A. Gers, "Learning to forget : Continual prediction with LSTM" 1999

8 E. L. Park, "KoNLPy : Korean natural language processing in Python" 6 : 2014

9 "KCDC"

10 D. C. Montgomery, "Introduction to linear regression analysis vol. 821" John Wiley & Sons 2012

11 A. Wilder-Smith, "Internet-based media coverage on dengue in Sri Lanka between 2007 and 2015" 9 (9): 31620-, 2016

12 R. Pascanu, "How to construct deep recurrent neural networks"

13 I. S. O. Hayate, "Forecasting word model : Twitter-based influenza surveillance and prediction" 76-86, 2016

14 S. Volkova, "Forecasting influenza-like illness dynamics for military populations using neural networks and social media" 12 (12): e0188941-, 2017

15 T. Mikolov, "Distributed representations of words and phrases and their compositionality" 3111-3119, 2013

16 C. de Almeida Marques-Toledo, "Dengue prediction by the web : Tweets are a useful tool for estimating and forecasting Dengue at country and city level" 11 (11): e0005729-, 2017