FolkSim: Analysis of Folksonomy Policies through Agent-based Simulation[go to overview]
The rapidly increasing popularity of Web 2.0 folksonomies originates in the ease of collaborative content creation and its sharing. However, the increasing scale of folksonomies poses a challenge on administration and moderation to ensure the quality of content and to prevent violations of laws (e.g. copyright protection, privacy, illegal content) and community rules. Understanding the interplay between community actors (users, moderators, administrators) and platform policies is crucial for the overall quality, user acceptance, and success of the entire folksonomy. In this paper we present our framework FolkSim which aims to analyze this complex problem through agent-based simulation. Our approach is instantiated for the popular scenario of collaborative question answering. As an example, we consider the real large-scale community of the Lycos IQ question answering platform. Results of our simulation experiments show a good fit with observed data. Systematic variation of simulated quality assurance mechanisms gives reasons for further design and optimization of quality-oriented community policies.
The FolkSim simlation framework is written in Java.
The source code can be download here.
Prof. Dr. Dr. Sergej Sizov
Prof. Dr. Steffen Staab