소셜 빅데이터 기반 보건복지 정책 미래신호 예측 = Future Signals of Health and Welfare Policies and Issues using Social Big Data
제목
소셜 빅데이터 기반 보건복지 정책 미래신호 예측 = Future Signals of Health and Welfare Policies and Issues using Social Big Data
저자
송태민
; 송주영
키워드
Social big data
; Machine learning
; Future signals
; Health&welfare
발행연도
2016-11-28
발행기관
한국보건정보통계학회
인용정보
보건정보통계학회지, vol. 41, no. 4, pp. 417 - 427
초록
Objectives: The purpose of this study is to collect health and welfare-related documents mentioned in and collectable from online channels, analyze important health and welfare keywords through topic and sentimental analyses, detect future signals concerning major policies and issues related to health and welfare services, and propose a prediction model.
Methods: 201,849 Health & Welfare related online documents from January 1 to March 31, 2016 from 171 Korean online channels and analyzed such documents using machine learning with random forest and Apriori algorithm association analysis. We used R software (version 3.2.1) for the association analysis data mining and visualization.
Results: As for the prediction of future signals of health and welfare policies, policies that were important and supported by the people were welfare payment, health promotion, job, marriage/child birth, health insurance, and healthcare industry (in this order). Specifically, as support for documents mentioning welfare payment and jobs was high, job creation through building a spontaneous welfare system is thought to be needed. Additionally, similar to the linkage analysis result of policies, as people were against documents that mentioned only {basic pension} policies, but supported documents that included {basic pension, welfare payment, job}, there is a strong demand for the establishment of a welfare system through active self-support and labor of the elderly.
Conclusions: Social big data can be utilized in various areas. First, similar to the application in this study, future signals concerning government`s policies and new technologies can be predicted in advance and prepared for. Second, they can be used as a new data collection methods that supplement limitations in survey data collection systems. Finally, a preemptive response system against risk can be established through monitoring and predicting social crisis.
