빅데이터 기반의 커뮤니티케어 키워드 및 사회연결망분석

Abstract

This study is aimed at identifying trends in research on community care and the knowledge structure of community care related research in a research paper published in the journal 'community care', which is the policy of the Ministry of Health and Welfare in 2018, and through network analysis and connectivity analysis using Word2vec, the research trends of 'community care' are improved. In this study, keyword extraction and analysis were conducted using word-clouding of the Python program in the journal Community Care, and the data were collected using Word2Vec of Deep Learning Technique (RNN) for the centrality and cohesion analysis of Semantic network analysis. The results of the study are as follows. First, the words frequently appearing through word-clouding were in the order of 'service', 'care', 'community', 'social', 'the aged', 'region', 'research', 'welfare', ' facilities', 'an analysis' and 'policy'. Second, 'Service, 'Community', 'Care', 'Social', 'the aged' and 'research' were evaluated as the most important nodes after visualizing the network using the connection strength between words obtained from Word2vec for 50 major keywords of the frequency of emergence. The words 'build', 'social', 'institutional', 'recuperation', 'region', 'United Kingdom', 'policy', insurance' and 'disabled' were found to be highly interconnected. Third, based on the network model's assessment of the centrality, 'social ', 'United Kingdom' 'disabled', 'policy' and 'Korea' were found to have a high centrality, the words 'treatment', 'Korea' service', 'the aged' and 'recuperation' and 'nearly' words 'social' and 'system', 'United Kingdom' were investigated. Finally, the groupings of research subjects using intuitive clustering over the network confirmed that they were clustered into three groups: the service sector, the treatment, the center, and the policy and the plan, and the aspects of community and welfare were revealed.