Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Master thesis: Can they agree: Searching for collaboration equilibria via multi-agent simulation of learning communities
3 views
Skip to first unread message
Zinayida Petrushyna
Jun 14, 2012, 4:56:39 AM6/14/12
to
Details: http://dbis.rwth-aachen.de/cms/can-they-agree-searching-for-collaboration-equilibria-via-multi-agent-simulation-of-learning-communities/view
What kind of persona are you in learning? Are you listening more than contributing to discussions - lurkers usually do, or do you initiate discussions and lead them - answering person or conversationalist behave like this. Dozens personas exist that are different in behaviors and goals during their learning processes. Using social media data and apply information retrieval techniques we were able to identify some of them[1].
Personas organize communities using social media. Structure of community networks depends on types of personas[2]. Changes in community structure are influenced from inside(community members) and outside world(incoming members). Goals of the members and strategies they use to achieve these goals impact on the structures as well. Members' strategies are not stable and changes over the time. Influences from the outside is not stable as well and can not be predicted. Therefore, we can not presume the changes in community based only on past data.
Multi-agent simulation is used to experiment with communities[3]. Using agent-based modeling we represent communities as agent-based systems that consist of a set of autonomous agents interacting with each other. They follow their strategies that changes because of inside or outside events in communities or just with no reason.
The topic of the thesis is to use given models of agents in the simulation framework (RePast, NetLogo, MATLAB). The student has to experiment with simulations to find simulation settings that reproduce community structure close to reality. The other task is to calculate collaborations equilibria[4] in a community where all collaboration participators comfortable enough not to change their collaboration strategies.
[1]Klamma, R., Spaniol, M., Denev, D.(2006):PALADIN: A Pattern Based Approach to Knowledge Discovery in Digital Social Networks, Published in 2006 by K. Tochtermann, H. Maurer (Eds.): Proceedings of I-KNOW '06, 6th International Conference on Knowledge Management, Graz, Austria, September 6 - 8, 2006, J.UCS (Journal of Universal Computer Science) Proceedings, Springer, pp. 457-464.
[2]Barabási, A.-L. , Albert, R.(1999): Emergence of scaling in random networks, Science (286), pp.509–512
[3]Gilbert, N. , Troitzsch, K.G.(2005): Simulation for the social scientist, McGraw-Hill International
[4]Dall’Astaa, L. , Marsilib, M., Pind, P.(2012): Collaboration in social networks, Proceedings of National Academy of Science of the United States of America
What kind of persona are you in learning? Are you listening more than contributing to discussions - lurkers usually do, or do you initiate discussions and lead them - answering person or conversationalist behave like this. Dozens personas exist that are different in behaviors and goals during their learning processes. Using social media data and apply information retrieval techniques we were able to identify some of them[1].
Personas organize communities using social media. Structure of community networks depends on types of personas[2]. Changes in community structure are influenced from inside(community members) and outside world(incoming members). Goals of the members and strategies they use to achieve these goals impact on the structures as well. Members' strategies are not stable and changes over the time. Influences from the outside is not stable as well and can not be predicted. Therefore, we can not presume the changes in community based only on past data.
Multi-agent simulation is used to experiment with communities[3]. Using agent-based modeling we represent communities as agent-based systems that consist of a set of autonomous agents interacting with each other. They follow their strategies that changes because of inside or outside events in communities or just with no reason.
The topic of the thesis is to use given models of agents in the simulation framework (RePast, NetLogo, MATLAB). The student has to experiment with simulations to find simulation settings that reproduce community structure close to reality. The other task is to calculate collaborations equilibria[4] in a community where all collaboration participators comfortable enough not to change their collaboration strategies.
[1]Klamma, R., Spaniol, M., Denev, D.(2006):PALADIN: A Pattern Based Approach to Knowledge Discovery in Digital Social Networks, Published in 2006 by K. Tochtermann, H. Maurer (Eds.): Proceedings of I-KNOW '06, 6th International Conference on Knowledge Management, Graz, Austria, September 6 - 8, 2006, J.UCS (Journal of Universal Computer Science) Proceedings, Springer, pp. 457-464.
[2]Barabási, A.-L. , Albert, R.(1999): Emergence of scaling in random networks, Science (286), pp.509–512
[3]Gilbert, N. , Troitzsch, K.G.(2005): Simulation for the social scientist, McGraw-Hill International
[4]Dall’Astaa, L. , Marsilib, M., Pind, P.(2012): Collaboration in social networks, Proceedings of National Academy of Science of the United States of America
0 new messages