分級醫療政策於社群媒體傳播與民眾參與

透過分級醫療議題在社群媒體上訊息發佈與時序性發展，探討社會大眾對於在社群媒體上關注健康政策的效用。利用臉書（Facebook）Graph API與人工搜尋方式，擷取2016年9月22日至2018年9月30日間，以內含「分級醫療六大政策」相關關鍵字的臉書公開貼文（posts）作為研究對象，觀察各貼文其發布來源，就發文數量與時序間的變化，以及民眾按讚、回應及分享產生之貼文民眾參與情形進行研究。並利用臺灣大學意見詞詞典繁體版（NTUSD-traditional）進行語意傾向之情感分析判斷。2016年9月22日至2018年9月30日間，對於分級醫療議題的發文，以個人帳號為最多，政府機關則是五類別中發文數量最低的。至於發文民眾參與，則以新聞媒體為最高，醫療機構則最低。情感分析結果，平均最正向的是政府機關，次之是醫療機構，接著由高至低分別為個人帳號、其他組織及新聞媒體。社群媒體提供可供民眾自由發表意見與看法的虛擬平台，有別於傳統瞭解民意的途徑，社群媒體在獲得訊息的同時，還能進一步分析抱持正向意見或反向意見的來源。因應社群媒體的興起，政府機關可評估將社群媒體納為瞭解多元民意的管道之一。

關鍵字

社群媒體；健康政策；分級醫療；民眾參與度；情感分析

並列摘要

This study investigated the amount of attention given by the public to information released via social media regarding the hierarchy of medical care policy and examined any change of attention with time. Facebook Graph API and manual searching were employed to collect public Facebook posts made between September 22, 2016 and September 30, 2018, containing keywords associated with the hierarchy of medical care policy. We observed the number of posts made by different categories in time sequence, and we also observed public engagement, which took the form of likes, comments, and sharing posts. Using the traditional Chinese version of the National Taiwan University Semantic Dictionary (NTUSD-traditional), we analyzed the semantic tendencies of posts. During the study period from September 22, 2016 to September 30, 2018, personal accounts made the most posts regarding the hierarchy of medical care policy, government agencies made the fewest posts. Regarding public engagement, news media posts received the most public attention, posts from medical institutes received the least public engagement. The sentiment analysis results revealed that government agencies posts were the most positive, followed by those from medical institutes, personal accounts, other organizations, and news media. The nature of social media is to provide a virtual platform where any user can freely express their own views and opinions. Unlike the traditional way of understanding public opinion, social media can further analyze the source of positive or negative information. Government agencies should thus consider adding social media as one of their means of understanding public opinion.

並列關鍵字

social media ； health policy ； the hierarchy of medical care ； public engagement ； sentiment analysis

參考文獻

張耕齊：Facebook作為社會劫盜地圖：意識形態估計、媒體偏斜、與輿情隔離。[臺灣大學經濟學研究所學位論文]，台北，2017。 doi: 10.6342/NTU201702253

陳百齡、鄭宇君、陳恭：社群媒體資料分析：特性和歷程的初探。傳播文化 2016;15:48-90。 doi: 10.30386/MCR.201407_(120).0004

Birnbaum ML, Garrett C, Baumel A, et al. Using digital media advertising in early psychosis intervention. Psychiatr Serv 2017;68:1144-9. doi: 10.1176/appi.ps.201600571

Utengen A, Rouholiman D, Gamble JG, et al. Patient participation at health care conferences: Engaged patients increase information flow, expand propagation, and deepen engagement in the conversation of tweets compared to physicians or researchers. J Med Internet Res 2017;19:e280. doi: 10.2196/jmir.8049

Jiang DD, Luo XF, Xuan JY, et al. Sentiment computing for the news event based on the social media big data. IEEE Access 2017;5:2373-82. doi: 10.1109/ACCESS.2016.2607218

主題瀏覽