Solving a Challenge for Users in Complex Systems

Understanding and interpreting the behaviors of others in an online environment is hard. The cues and signals that we readily interpret in a face-to-face situation are not present or at best attenuated. Lacking sufficient cues, users often misinterpret or misunderstand the actions and intentions of others. As the number of participants and the amount of interaction grows it becomes harder and harder for users to make sense of others actions, much less their own place in the community and the health of the community at large.


Through several projects we are exploring how to support social translucence in collaborations that involve large numbers of participants. Below we highlight some current and prior projects.



Translucence Support with Scalable Machine Learning

Understanding the different types of contributions that users make is important to providing social translucence. In this work we use machine learning to transform one type of behavioral cue into another that is more salient. Specifically, we take the practice of awarding barnstars, small tokens that acknowledge and appreciate different forms of work in Wikipedia, and shift the focus from the explicit action of an edit that awards one single barnstar, to the more salient action of recognizing contributions by the individual who receives barnstars. We are applying machine learning to a multi-label classification problem to identify the types of work that individuals contribute as identified through their cumulative barnstars. Because of the high-dimensionality of the dataset, we are working on distributed versions of traditional machine learning algorithms for multi-label classification.


Related Publications

Sajani, H., S. Javanmardi, D. W. McDonald and C. V. Lopes. (2011) Multi-Label Classification of Short Text: A Study on Wikipedia Barnstars. Workshop on Short Text Classification (AAAI-11). -- to appear


Kriplean, T., I. Beschastnikh, and D. W. McDonald. (2008) Articulations of Wikiwork: Uncovering Valued Work in Wikipedia through Barnstars. Proceedings of the 2008 ACM Conference on Computer Supported Cooperative Work (CSCW’08).



Permission Granting and User Promotions

Decision making in an online community is often a difficult process. In large online communities like Wikipedia, administrators are key to keeping the community functioning, and those administrators come from the ranks of regular users. The decision to grant administrative permissions has important consequences for the entire community. In the process of deciding which candidates should be granted administrative permissions, reviewers must consider a user’s behavior relative to a set of criteria and come to some reasonably shared understanding of the merits of a new potential administrator. This work examines the collaborative activity of an online community deciding who is given administrative privileges.


Related Publications

Derthick, K., P. Tsao, T. Kriplean, A. Borning, M. Zachry, and D. W. McDonald (2011) Collaborative Sensemaking during Admin Permission Granting in Wikipedia. In Proceedings of the HCI International (HCII'11).



Self-Presentation through Semi-Structured Profiles

The study of self-presentation in online social networks has relied on data collected from structured profiles. However, structured profiles may present a biased view by influencing the type of information users express. This project considers how Wikipedia userboxes are used as an alternative form of self-presentation. Overall, we seek to understand how reducing profile structure changes the nature of online presentation and the impact of that change on supporting socially translucent interaction


Related Publications

Le, L., I. Beschastnikh, and D. W. McDonald (2010) Self-Presentation: Structured and semi-structured user profiles. Presented at the “Models, theories and methods of studying online behavior” Workshop at the ACM 2010 SIGCHI Conference on Human Factors in Computing Systems (CHI'10).



Consensus and Self Governance in Wikipedia

When large groups cooperate, issues of conflict and control surface because of differences in perspective. Managing such diverse views is a persistent problem in cooperative group work. The Wikipedian community has responded with an evolving body of policies that provide shared principles, processes, and strategies for collaboration.


We focus on the enactment of policies in discussions on the talk pages that accompany each article. These policy citations are a valuable micro-to-macro connection between everyday action, communal norms and the governance structure of Wikipedia. We find that policies are widely used by registered users and administrators, that their use is converging and stabilizing in and across these groups, and that their use illustrates the growing importance of certain types of less visible and less obvious work that sustains large collaborative activity.


Related Publications

Beschastnikh, I., T. Kriplean, and D. W. McDonald (2008) Wikipedian Self-Governance in Action: Motivating the Policy Lens. Proceedings of the Second International Conference on Weblogs and Social Media (ICWSM'08). Seattle, Washington.


Kriplean, T., I. Beschastnikh, D. W. McDonald and S. Golder. (2007) Community, Consensus, Coercion, Control: CS*W or How Policy Mediates Mass Participation. Proceedings of the ACM 2007 International Conference on Supporting Group Work (GROUP'07).



Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant No. IIS-0811210. Additional support provided by Amazon.com and Overland Storage. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.