Structure and Dynamics of Collaborative Information Spaces
This project is related to our research line: Design and analysis of socio-technical systems.
Duration 36 months (May 2015 - April 2018)
Funding source Swiss State Secretariat for Education, Research and Innovation (SERI) (Project C14.0036)
This project is related to the EU COST Action TD1210: "Analyzing the dynamics of information and knowledge landscapes - KNOWeSCAPE"
The convergence of social and technical systems raises a number of important and novel issues. Knowledge spaces like, e.g., the WWW are created, organised and consumed in an increasingly collaborative fashion by groups of humans interacting on short time scales, a process commonly subsumed under the umbrella of social computing or social information processing.
As such, the question how pieces of information are linked to each other, ranked and filtered not only affects the ability of individuals or organisations to access information in a timely, objective and transparent manner. It is also of prime importance for society as a whole since notions of relevance in networks of linked information a) are increasingly influenced by social processes and b) can be an important driver of social dynamics themselves. The resulting feedback between the social and the semantic layer of collaborative knowledge spaces questions to what extent current network-based information ranking measures - even though they are computed by algorithms - can actually be seen as objective. Although the social and the information layer of collaborative knowledge spaces are coupled inseparably, the question how knowledge orders and social dynamics influence each other has been addressed only partially so far. A systems perspective that integrates both layers is still missing.
Figure: In this project, we will explore the possibility to utilise second-order time-aggregated representations (see above) of human navigation behavior for the ranking of information.
In this project, we close this gap by studying social feedback phenomena in information networks from the perspective of multiplex and dynamic networks. By taking a multiplex network perspective, we consider both the social and the information layer of collaborative knowledge spaces, and quantify their properties as well as mutual dependence. Complementary, by employing our competence in temporal network theory, we quantify the navigation behavior of users within information spaces from a complex networks perspective. We study how the retrievability of content and its ranking, in terms of predominant measures of relevance, is influenced by the structure and dynamics of the social systems that create it. We consider this question to be of particular interest and significance due to the recent trend towards "socially influenced" information retrieval systems, like e.g. social search or collaborative filtering techniques. We further address the question how increasingly accessible data on the social dimension of collaborative knowledge spaces can be used to augment existing relevance ranking mechanisms.