The Web Science Trust

Studying Forum Dynamics from a User Engagement Perspective

Kan, Andrey and Chan, Jeffrey and Hayes, Conor and Hogan, Bernie (2011) Studying Forum Dynamics from a User Engagement Perspective. pp. 1-3. In: Proceedings of the ACM WebSci'11, June 14-17 2011, Koblenz, Germany.

PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Image (PNG)

Official URL:


Online forums are popular websites that provide extensive communication records of users over time. Systematic anal- ysis of these records can facilitate the mathematical model- ing of communication and the development of forum man- agement tools. For example, the models can be capable of prediction whether a user is likely to elicit responses in the future, based on the past observations. Many researchers have tried dierent approaches to forum data analysis, such as a focus on user roles, discussion diver- sity and growth dynamics. For example, Both Viegas and Smith [4] and Welser et al. [5] use visualization to derive structural features of the roles of individual users thus indi- cating clear dierences between consistent \answer people" and those preferring discussions. Chan and Hayes [1] extend this work by categorizing such roles and exploring the com- position of forums in terms of a diversity of roles. Kumar et al. [3] and Gonzalez-Bailon et al., [2] model growing trees of discussion threads. This previous work has proven germane, but is constrained in several ways. First, these studies tend to assume that users exhibit consistent behavior over time. Second, studies that look to temporal patterns over time instead of assuming consistency (such as AuthorLines [4]) tend to be descriptive, visual and dicult to scale. To address these limitations, we propose a method to rep- resent a user's engagement in communication over time as a user path that is a line in a space of user features.

Item Type:Conference or Workshop Item (Poster)
Web Science Comments:WebSci Conference 2011
Subjects:WS1 Computer Science
WS3 Web Engineering
Web Science Events > Web Science 2011
ID Code:459
Deposited By: Lisa Sugiura
Deposited On:07 Jun 2011 16:43
Last Modified:25 Oct 2011 17:12

Repository Staff Only: item control page

EPrints Logo
Web Science Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.