This project proposes to leverage massive scale Web 2.0 data for the purpose of automating the identification of design principles for constructing Web 3.0 where online communities serve the dual purpose of learning and work occurring in synergy. Web 2.0 has produced ample evidence that through careful integration of theories with powerful architectures for supporting massive scale communication for the purposes of deliberation, innovation, consensus building, and coordination of action, we can engage the masses in open production communities to architect resources such as Wikipedia, GitHub, and Crowdsourcing Communities like Amazon’s Mechanical Turk. In the same way, networked learning, where the recent phenomenon of Massive Open Online Courses (MOOCs) is one example, and the well established Knowledge Building Communities in connection with the Knowledge Forum is another, pose a corresponding form of online participation where communities of shorter duration form for the purpose of providing a context in which individuals can better themselves. In all cases, the greatest resource lies in the great diversity of perspectives, expertise, skills, and energy of the masses. The data traces from these endeavors provide a valuable resource towards the design of Web 3.0.
What motivates the proposed work is the observation that despite the tremendous potential, the massive resource in crowd power of online communities comes with equally great challenges. Decades of research studying the inner workings of these communities, whether focused on learning, work or a hybrid of the two, has revealed that a careful balance of exploration, consensus building, and coordination do not occur naturally without support. The proposed research will produce design principles and powerful modeling frameworks for development of intelligent support for participation that achieves a careful balance of guidance and self-direction by leveraging existing data from Web 2.0. In particular, a focus of the work will be on modeling the individual differences of participants so that a diverse population of potential participants are equally well supported. In that way, the hope is to leverage not only the typical internet participant of Web 2.0, but to broaden participation in Web 3.0 to accommodate a more diverse population of participants including the underserved.
This proposal brings together leading researchers in the areas of Language Technologies, Human-Computer Interaction, Computer Supported Cooperative Work, Computer Supported Collaborative Learning, and Education to attack this problem with a multidisciplinary three pronged approach. In particular, the proposed research program is to: (1) Lay a foundation in a versatile data infrastructure that will enable integration of heterogenous data sources related to discussion and work coordination; (2) Build and interpret models of group and community interactions in learning and work settings that link processes with outcomes in order to identify sets of functional behavior profiles that together form the foundation for support of on boarding, participation, productivity, and mentoring; and (3) Apply interpretation of models to empirically grounded principles that motivate the design of online communities that are conducive to learning and work.