English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 80990/80990 (100%)
造訪人次 : 41267827      線上人數 : 97
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋


    請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/93085


    題名: 以 Reddit 使用者生成內容探討糖尿病照護社會支持;Social Support for Diabetes Care on Reddit: A User-Generated Content Approach
    作者: 洪宗銘;Hung, Tsung-Ming
    貢獻者: 資訊管理學系
    關鍵詞: 糖尿病;社群媒體;社會支持;Reddit;詞嵌入;深度學習;Diabetes;social media;social support;Reddit;word embedding;deep learning
    日期: 2023-06-29
    上傳時間: 2024-09-19 16:41:24 (UTC+8)
    出版者: 國立中央大學
    摘要: 隨著全球糖尿病盛行率的持續增長以及社群媒體在慢性疾病管理中的重要性日益 凸顯,理解糖尿病相關的線上社群支持行為模式變得至關重要。本研究聚焦於 Reddit 糖 尿病社群,對使用者在該社群中尋求與提供支持的行為模式進行深入探討。通過分析 466 則貼文和 2,546 條回覆,並應用詞嵌入技術與深度學習分類模型,對貼文中的支持尋求 策略和回覆中的社會支持類型進行了自動化分類。結果顯示,社群成員在尋求支持時主 要採取「尋求支持」策略,而在提供支持時以「資訊性支持」為主。同時,研究也發現, 不同尋求策略會對應到不同的社會支持類別,如「慶祝成就」的策略大多獲得「尊重支 持」。這些發現對於糖尿病護者及其醫療保健人員提供了寶貴的行為指引,有助於提供 更精準的支持與指導。使外,在模型方面,貼文分類模型的準確率達到了 90%,而回覆 分類模型的準確率達 72%。這些結果為社群研究者和資料科學家在自動化分類支持尋求 策略和社會支持類別提供了有效的參數及方法。;As the global prevalence of diabetes continues to rise and the significance of social media in the management of chronic diseases increasingly stands out, it has become critical to understand the online community support behavior patterns associated with diabetes. This study focuses on the Reddit diabetes community, exploring in-depth the behavioral patterns of users seeking and providing support within this community. Through analyzing 466 posts and 2,546 comments, and applying word embedding techniques and deep learning classification models, the study automatically classifies support seeking strategies in posts and types of social support in comments. Results indicate that when community members seek support, they mainly adopt the "Seeking support" strategy, whereas "Informational support" is primarily provided. Furthermore, it was discovered that different seeking strategies correspond to different categories of social support, such as the "Celebrating achievements" strategy mostly receiving "Esteem support". These findings provide valuable behavioral guidance for diabetes caregivers and their healthcare professionals, aiding in delivering more precise support and guidance. Regarding the models, the post classification model achieved an accuracy of 90%, while the comment classification model reached an accuracy of 72%. These results offer effective parameters and methods for community researchers and data scientists in automatically classifying support-seeking strategies and social support categories.
    顯示於類別:[資訊管理研究所] 博碩士論文

    文件中的檔案:

    檔案 描述 大小格式瀏覽次數
    index.html0KbHTML9檢視/開啟


    在NCUIR中所有的資料項目都受到原著作權保護.

    社群 sharing

    ::: Copyright National Central University. | 國立中央大學圖書館版權所有 | 收藏本站 | 設為首頁 | 最佳瀏覽畫面: 1024*768 | 建站日期:8-24-2009 :::
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 隱私權政策聲明