Workshops
Implicit Online Communities
Length: half day
Definitions of community have taken on new meanings facilitated by the advancement of Information and Communication Technology (ICT). Traditionally, the focus of online community research has been on message-based communications between users who share some social context in common. The common element may be geography, needs, practices, organizations, or other bases for social connection. The analysis is mainly done around messages and the social characteristics of the users who post them. As ICTs advance, they enable new, implicit forms of communications that take advantage of the low communication costs. Thanks to new technologies such as online content sharing (e.g., YouTube, Flickr), voting (Digg.com), tagging and folksonomy (del.icio.us and WikiMapia), and social search (Jookster), non-traditional communities are built to aggregate user activities or preferences where users do not have devote as much attention or effort to the community to participate as in traditional online communications such as in forums and bulletin boards. These new communities are special in that there is not much traditional online communication at all, as users communicate through observing other users' actions instead of messages. For example, when a user visits YouTube, his action of viewing a particular video contributes to the "total user views" result and will be displayed in the video's web page. Other users visiting YouTube will get to see the list of most viewed videos and their viewing behavior may be affected. Yet, there is little understanding of such new implicit community phenomena. Most of the extant relevant work is on collaborative filtering and technical or analytical in nature--there is no empirical investigation on how if any influence of such kind of recommendation works or not. We believe that personal influences through tacit recommendation and actions do exist. Such influences, however subtle, may have non-trivial consequences in shaping users' preferences and decisions.
Fortunately, the online nature of the communities means that data can be collected to analyze user behavior. With the vast amount of data potentially available to researchers, especially user activity data, opportunities abound for researchers to analyze the patterns of user activities and how users are affected by each other. As such data become abundant for researchers, there are many theoretical issues that need to be resolved before one can engage in analyzing the activities data. For example, an inherent issue in studying collaborative filtering empirically is how to accurately pinpoint its impact on user actions out of other possible exogenous reasons, considering the heterogeneity of users and the spontaneous nature of their online actions.
This workshop aims to discuss research topics and issues in new implicit communities. Workshop participants are requested to submit abstracts and/or papers Mu Xia at mxia AT uiuc.edu by May 8, 2007 (note: new submission date). We will notify participants of the status of their submissions by May 20, 2007. Abstracts and papers can be work in progress or completed work, on such themes as:
- Empirical studies on implicit recommendation
- Implicit community data collection techniques and issues
- Theory development for implicit communities.
- Tools to visualize and analyze implicit communities
An informal review process will be used to ensure the quality.
The workshop will conclude with a panel discussion, with experts from both industry and academia discussing promises and challenges in studying new technology-enabled communities.
The Organizers
Mu Xia is an assistant professor of Information Systems in the College of Business at the University of Illinois at Urbana-Champaign.
Wenjing (Wendy) Duan is an assistant professor in the Dept. of Information Systems and Technology Management in the School of Business at George Washington University.