【2016学术报告05】Online Community Oriented Data Analytics
Title: Online Community Oriented Data Analytics
Speaker: Bin Zhang, Assistant Professor, University of Arizona
Time: 10:30-12:00, 29th, Jun. (Fri.)
Place: 1-315, FIT Building
Organizer: Research Institute of Information Technology (RIIT), Tsinghua University
Bin Zhang is an assistant professor in the Department of Management Information Systems, University of Arizona (UA), and a visiting research fellow at Carnegie Mellon University (CMU). He is also an affiliated member of UA’s Artificial Intelligence Lab. Bin received his Ph.D. degree in Information Systems Management from CMU, and a Master's degree in Machine Learning, from CMU's School of Computer Science. His primary research interests are large social network analysis and statistical modeling for network problems. Bin's research projects have been funded by federal agencies such as NSF and NIH. His work has appeared in premier information systems journals and conferences. Bin also has experience in the Internet industry at companies like Yahoo! and has designed architectures of online ERP systems in the software industry.
Two types of online communities are addressed in this talk. In the first part, online doctor consultation services (ODCS), a new type of online health communities (OHC), is studied. ODCS enables doctors deliver healthcare directly to patients by consultation. Patients can use both online attributes, such as reviews, and offline attributes, such as position rank, of doctors to choose the consultation provider. However, it is uncertain whether the online review is really used by patients, and existing literature shows contradictory evidence. This paper attempts to reconcile such problem and identify factors affecting patients’ decision. Results show that online review does affect patients choose doctor, although offline attributes are also considered. Online review can weaken the effect of doctor’s hospital tier level and the effect of position rank is insignificant, suggesting patients care more about online review than the doctor’s position and hospital. Our findings can help doctors design their strategy of online service, also help ODCS refine their system design so that consultation matching is optimized.
The second part of the talk is about virtual inter-organizational community of practice (IOCoP). It enables professionals to exchange and share knowledge via computer-mediated interactions. Prior literature mainly focuses on internal motivating factors at the individual level. However, knowledge sharing requires social interactions and influences from external entities play an important role in encouraging individuals’ community participation. In this research, we study external motivating factors generated from two channels: peer effects and organizational influences outside the virtual community. We apply a novel identification method to analyze a unique dataset in the financial trading sector. We find that external motivating factors from peers and organizations are influential in determining community participation quantitatively as well as qualitatively. Differentiating motivating factors enables us to design various mechanisms with which IOCoPs can apply to engage collective learning and knowledge management across organizations.