SPEAKER
Pr William Rand
North Carolina State University
DATE
26/06/2018
- Video / Presentation
To watch in full screen mode, start the video and click on the "UNS POD" logo- Abstract
- Social media provides a platform, where firms can interact directly with individual consumers. Firms often cannot and sometimes should not respond to every user. When a consumer mentions a company on the social media platform the firm has to decide whether or not to respond to that specific individual. Knowing as much as possible about the user, is vital in order to maximize the firm’s internet customer relationship management. The user may clearly provide the necessary information, but how can this information be elicited when the user is less forthcoming. Three desired characteristics are being investigated: geography, customer lifetime value, and internet word of mouth influence. Casual State Models (CSM) are able to model a user’s social media behavior. Through modeling the behaviors of users with the known desired and undesired characteristics, we are able to test whether the differences in the CSMs is enough to classify an unknown user. CSMs are built on the individual and group levels to determine whether it is necessary to model the individual completely, or whether they can be safely modeled as just another member of the group. The individual models performed better than the group levels for classifying the unknown users. Expanding the social media behavior to include more features may help improve the current models.
- About the speaker
- William Rand examines the use of computational modeling techniques, such as agent-based modeling, machine learning, network analysis, natural language processing, and geographic information systems, to help understand and analyze complex systems, such as the diffusion of information, organizational learning, and economic markets. He also works to develop methods, create pedagogy, and build frameworks to allow researchers and practitioners to use analytics and data-intensive methods in their own work. He has received funding for his research from the NSF, DARPA, ARL, Google, WPP, and the Marketing Science Institute. He received his doctorate in Computer Science from the University of Michigan in 2005 and prior to coming to NCSU was at the University of Maryland for eight years.