讲座题目：Matching Donors to Projects on Charitable Giving Platforms
地点： 深圳大学文科楼 管理学院资料室2423D 10:20am-12:00pm
Matching donors with causes they would like to support is critically important in philanthropy. Typically, donors are not willing to incur significant search cost to find projects that match their interest. Therefore, historically managers in non-profits and community foundations have carried out the matching based on their knowledge of donors’ interests and the agendas of the philanthropic organizations. However, the recent advances in online peer-to-peer fundraising has the potential to fundamentally change the matching process. Donors are increasingly turning to online charitable giving platforms to access a large number of charities, and vice versa. This has led to the aggregation of large amount of fundraising and donation data at these platforms. Therefore, instead of relying on the ability of individual match-makers, there is now an opportunity to take a data-driven approach to understand donor’s preferences and match with her, among large number of projects, the few that are most suitable for her.
We propose an approach to match donors to projects on online charitable giving platforms, taking into account donors’ preferences, budget, and cognitive limitations as well as the dynamic status and budgetary needs of the projects. Our approach is based on a structural model of donors’ behavior driven by the impure altruism from philanthropic acts. We explicitly model the time taken by a donor to return to the platform after a donation as a result of her exposure to fundraising campaigns and prior funding experience, the effect of the small latent subset of projects a donor might be aware of out of thousands of available projects at any time, and the utility maximizing allocation of a donor’s budget to donations. We estimate the proposed model using a highly granular donations and fundraising campaigns dataset from DonorsChoose. We demonstrate that the proposed model better captures donation behavior than several benchmarks using a set of model fit measures and out-of-sample prediction. Using the estimated model, we design and evaluate optimal recommendation policies to maximize fundraising success. By matching projects to donors, not only based on the donors’ preference, but also their budget, and their willingness to support projects with different odds of success; as well as the needs of the projects, the optimal recommendation strategies increase the donations raised by about 22% from the current levels.
Dr. Zhuoxin Li received his Ph.D. in Information, Risk, and Operations Management from the University of Texas at Austin in 2015. He has also received his M.S. in Management Science from Harbin Institute of Technology and B.S. in Computer Science from South China University of Technology in 2009 and 2007, respectively. He joined the Carroll School of Management at Boston College as an Assistant Professor in 2015. Dr. Li’s research focuses on the information economics, online platforms and electronic commerce. He been published in Management Science and Production and Operations Management.
沙发#作者：游客 发布于：2018-06-27 22:38