Seminar (Dept. of Marketing)

Title: Recommendation Algorithms on User-Generated Content in Social Networks: the Quantity-Quality Dilemma

Speaker: Ziwei Cong( McDonough School of Business, Georgetown University.)

Time: 10:00-11:30am, June 5th (Wednesday), 2024

Venue：Room 1008, Mingde Business Building

Language：English

Abstract：

This paper examines the role of recommendation algorithms in shaping user-generated content (UGC) within online social networks, using a quasi-experiment on a large online Q&A community platform. Originally relying on a content-based filtering algorithm that recommends content based on user subscribed topics, the platform completely moved to a social filtering algorithm, which recommends content with which users’ social connections are engaged. Our analysis reveals that this shift significantly improved the user-perceived quality of UGC by 5%, at the cost of a significant reduction in UGC quantity. There was a 18% reduction in answer contribution, a 45% reduction in answer number per question, and a 555% increase in the response time for the first answer. We further explore the mechanisms for this quantity-quality trade-off by comparing the platform’s social and topic subscriber networks, in terms of content distribution and user homophily. We document that while topic subscriber networks provide much broader content reach, social networks more effectively gather users with similar interests, enhancing content relevance and targeting.

Short biography：

Ziwei Cong is an Assistant Professor of Marketing at McDonough School of Business, Georgetown University. Her research areas include creator economy, influencer marketing, and platform designs. Her research combines methods from statistics, econometrics, and machine learning, and has been published in leading Marketing journals (e.g., Journal of Marketing Research). She is the winner of the 2021 Vithala R. and Saroj V. Rao ISMS Doctoral Dissertation Award, the runner-up of the 2020 Shankar-Spiegel Dissertation Proposal Award, and the finalist of the 2020 Best Doctoral Dissertation Proposal Competition at the American Statistics Association (Marketing Section). She has served as an ad-hoc reviewer at Marketing Science and Management Science.