Publications-Proceedings
Article View/Open
Publication Export
-
Google ScholarTM
NCCU Library
Citation Infomation
Related Publications in TAIR
題名 Unveiling the Power of Interaction: Key to Charitable Fundraising Success on GoFundMe 作者 向倩儀
Hsiang, Chien-Yi貢獻者 資管系 關鍵詞 Charitable Fundraising; Feature Engineering; Natural Language Processing; Machine Learning; Interaction 日期 2024-12 上傳時間 14-Feb-2025 10:47:03 (UTC+8) 摘要 This study investigates the factors influencing the success of charitable fundraising campaigns on GoFundMe, emphasizing the role of interaction features. Six machine learning models (KNN, Logistic Regression, SVM, ANN, Decision Tree, AdaBoost) were used to evaluate feature predictive power. Combining interaction features with project information significantly enhanced accuracy, with Logistic Regression and AdaBoost achieving perfect AUC scores of 1.00. Decision Tree results across 18 project categories demonstrated that integrating interaction features improved predictive outcomes, especially for Business, Charity, and Memorial fundraisers, achieving AUC scores of 0.86, 0.81, and 0.88, respectively. These findings highlight the importance of donor interactions in boosting campaign success. This research provides actionable insights for charitable organizations to optimize their fundraising strategies and increase donor engagement, ultimately maximizing the impact of their campaigns. 關聯 ACIS 2024 conference proceedings, Australasian Association for Information Systems 資料類型 conference dc.contributor 資管系 dc.creator (作者) 向倩儀 dc.creator (作者) Hsiang, Chien-Yi dc.date (日期) 2024-12 dc.date.accessioned 14-Feb-2025 10:47:03 (UTC+8) - dc.date.available 14-Feb-2025 10:47:03 (UTC+8) - dc.date.issued (上傳時間) 14-Feb-2025 10:47:03 (UTC+8) - dc.identifier.uri (URI) https://nccur.lib.nccu.edu.tw/handle/140.119/155690 - dc.description.abstract (摘要) This study investigates the factors influencing the success of charitable fundraising campaigns on GoFundMe, emphasizing the role of interaction features. Six machine learning models (KNN, Logistic Regression, SVM, ANN, Decision Tree, AdaBoost) were used to evaluate feature predictive power. Combining interaction features with project information significantly enhanced accuracy, with Logistic Regression and AdaBoost achieving perfect AUC scores of 1.00. Decision Tree results across 18 project categories demonstrated that integrating interaction features improved predictive outcomes, especially for Business, Charity, and Memorial fundraisers, achieving AUC scores of 0.86, 0.81, and 0.88, respectively. These findings highlight the importance of donor interactions in boosting campaign success. This research provides actionable insights for charitable organizations to optimize their fundraising strategies and increase donor engagement, ultimately maximizing the impact of their campaigns. dc.format.extent 102 bytes - dc.format.mimetype text/html - dc.relation (關聯) ACIS 2024 conference proceedings, Australasian Association for Information Systems dc.subject (關鍵詞) Charitable Fundraising; Feature Engineering; Natural Language Processing; Machine Learning; Interaction dc.title (題名) Unveiling the Power of Interaction: Key to Charitable Fundraising Success on GoFundMe dc.type (資料類型) conference
