• Home
  • Introduction
  • Faculty Directory
  • Seminar
  • Contact Us
  • Peking University
  • Belt & Road Institute
  • 中文
  • Seminar

    speaker Zhiqiang (Eric) Zheng, University of Texas at Dallas time Monday, June 24th , 09:00-10:30
    place Room 217, Guanghua Building 2

    Management Science and Information Systems' Seminar(2019-14)

    Title: Detecting and Deterring Online Customer Misbehavior

    Speaker: Zhiqiang (Eric) Zheng,University of Texas at Dallas

    Time: Monday, June 24th, 09:00-10:30

    Location: Room 217, Guanghua Building 2


    Customer misbehavior in firms’social media presents a serious challenge to businesses. This paper develops an algorithmic approach to detect and restrain online customer misbehavior. Specifically, we build on natural language processing and deep learning techniques to automatically detect customer misbehaviors by mining a firm’s social media data, and then implement two norm enforcement mechanisms– punishment and common identity –to restrain customer misbehavior without compromising customer relationship. In collaboration with a prominent apparel e-commerce company, we conducted a field experiment based on the unique data provided by the firm. The results show that our algorithmic solution achieves excellent performance in detecting customer misbehavior, improving detection performance by 7 to 9 percent on average. The results from the field experiment indicate that punishment reduces customer misbehavior considerably in the short term, but this effect decays over time; whereas common identity has an initially small but persistent effect on misbehavior reduction in the long run. Importantly, we show that the combination of punishment and common identity can achieve a more substantial and longer-term reduction. In addition, we find that punishing dysfunctional customers decreases their purchase frequency, whereas imposing a common identity on those customers improves their purchase behavior. Interestingly, our results show that combining the two enforcement mechanisms offers an effective control strategy to alleviate the detrimental effect of punishment. Finally, we reveal the heterogeneous treatment effect on customers ranging from novice to experienced ones.


    Zhiqiang (Eric) Zhengis the Ashbel Smith Professor of Information Systems with a joint appointment in Finance at the Jindal School of Management, University of Texas at Dallas. He received his PhD from the Wharton School of Business, University of Pennsylvania. His current research interests focus on financial technology, Blockchain analytics and healthcare analytics. His papers have appeared in journals such as Information Systems Research, Management Science, MIS Quarterly, and Production and Operations Management. He currently serves as a senior editor for Information Systems Research and is the editor for the Fintech and Blockchain special issue.

    Your participation is warmly welcomed!

    © 2019 Guanghua School of Management Peking University