Ph.D. Position: temporal and semantic analysis of richly typed social networks from user-generated-content sites on the web
Title: temporal and semantic analysis of richly typed social networks from user-generated-content sites on the web
Domain: Perception, cognition, interaction
About the team: The Wimmics stands for Web-Instrumented Man-Machine Interactions, Communities, and Semantics. Wimmics research team conducts research in the domain of graph-based knowledge representation and inference systems applied in particular to semantic web and social web. The research fields of this team are graph-oriented knowledge representation, reasoning and operationalization to model and support actors, actions and interactions in web-based epistemic communities.
Topic: There is now a strong interest in combining Social Network Analysis with Semantic Web frameworks in order to include inferences in the analysis of the social structure and exchanged data we capture through online social web sites.
But in many scenarios we not only need to understand the structure of an epistemic community and its resources, we also need to monitor the evolutions, trends, flows, signals, etc. of the dynamics of this complex networking.
In this Ph.D. proposal the candidate will consider how we can reconcile temporal reasoning, social network analysis, semantic web formalisms and graph dynamics to provide an innovative conceptual framework for temporal and semantic analysis of richly typed social networks from user-generated-content sites on the web.
The work plan includes:
- Study of the state of the art of models and algorithms to formalize and semantically analyze folksonomies, forum, discussions and social networks in the context of communities of interest on the open web.
- Exploit these representations to support the detection of possibly overlapping communities of interest, the roles and characteristics of users in the community, their profiles, and their interactions and relations.
- Revisit these semantic and structural models to integrate temporality in their formal definitions to go beyond the static analysis of a snapshot of the network and support the detection of emerging communities and dying ones, new topics, recurrent long term interests, etc.
An important aspect of this work will be the integration of the time dimension to semantic web frameworks for the representations of social web application data and the extension of semantic web operators (e.g. SPARQL), the social network metrics (e.g. betweenness centrality) and community analysis algorithms (e.g. labeled community detection) to exploit the temporal dimension and provide trends indicators.
Profile: The student should have a master degree in computer science and a strong knowledge of web and preferably semantic web standards.
Duration: 36 months
Place: Inria Sophia Antipolis
Team page: http://wimmics.inria.fr