题　目：From Data to Donations: Optimal Fundraising Campaigns for Non-Profit Organizations

嘉　宾：Zhengchao Wang, Ph.D. Candidate, Imperial College London

主持人：花成 副教授 金沙威尼斯欢乐娱人城

时　间：2024年11月29日（周五）10:00-11:30am

地　点：安泰楼A503室

内容简介：

Non-profit organizations play an essential role in addressing global challenges, yet their financial sustainability often depends on raising funds through resource-intensive campaigns. We partner with a major international non-profit to develop and test data-driven approaches to enhance the efficiency of their fundraising efforts. The organization conducts multiple annually recurring campaigns, with thematic links between them. These connections enable us to predict a donor’s propensity to contribute to one campaign based on their behavior in others. This structure is common among non-profits but not readily utilized by conventional multi-armed bandit algorithms. To leverage these inter-campaign patterns, we design two algorithms that integrate concepts from the multi-armed bandit and clustering literature. We analyze the theoretical properties of both algorithms, and we empirically validate their effectiveness on both synthetic and real data from our partner organization.

演讲人简介：

Wang Zhengchao is a fifth-year PhD candidate at Imperial College Business School, working with Professor Wolfram Wiesemann and Professor Heikki Peura. His research centers on data-driven decision-making, with a focus on leveraging robust optimization and machine learning to address challenges in operations management. His work spans diverse domains, including revenue management, sustainable energy system management, and the operational management of non-profit organizations. Zhengchao's research has been published in prestigious journals such as Operations Research. In addition to his academic contributions, he serves as a reviewer for leading journals, including Operations Research, Operations Research Letters, and Applied Energy, as well as for top artificial intelligence conferences such as NeurIPS, ICML, and ICLR.

欢迎广大师生参加！