[Corpora-List] Funded PhD (in Paris area): Modeling the socio-cognitive dynamics of digital communities

[Thierry Poibeau]

PhD Scholarship

*** Modeling the socio-cognitive dynamics of digital communities ***

LaTTiCe-CNRS and ISC-PIF offer a PhD scholarship on the analysis of socio-cognitive dynamics in digital communities on theWeb. The scholarship is sponsored by the Region Ile de France and is hosted by LaTTiCe-CNRS (http://www.lattice.cnrs.fr/) and ISCPIF (Complex System Institute, http://www.iscpif.fr/). Both laboratories are located in the Paris area.

- Deadline for submitting an application : 17 July 2011,
- Grant starting on : October or November 2011,
- Duration : three years,
- Salary : 1 450 euros a month
- Contacts : Thierry Poibeau (firstname.lastname at ens.fr) and Jean-Philippe Cointet (firstname.lastname at polytechnique.org)

With the development of the Web, the production and circulation of knowledge faces a major change. Information can now be found in digital archives widely distributed and easily accessible across the planet. This new landscape, although built by the activity of individuals, has its own dynamics. We think this context makes it possible to observe the
emergence of a number of specific but persistent semantic structures.

*** Objectives ***

The goal of this research is to better understand the dynamics of online communities [2]. The idea is to work on the dynamics of ideas (through a thorough analysis of the linguistic content produced by these communities ) as well as on the dynamics of the communities themselves (through an analysis of the social network behind the communities) [4]. In order to do so, it is necessary to develop realistic dynamic models of the development of these communities. It may thus be possible to discover some of the active processes at the origin of this development.

The successful candidate will more specifically tackle the following research issues :

1. the reconstruction of the mesoscopic dynamics of these communities. In a community of bloggers, how do new topics emerge? How do they evolve? How to model “influencal actors”? How does the social structure evolve with time? What are the relations between the communities and between the actors themselves?
2. the comprehension of the distribution and movement of entities in these communities. The goal is to model the dynamics of content along with the dynamics of actors, taking into account the constraints imposed by both high-level structures : the social and the semantic networks [1].
3. the analysis of the appearance of new ideas or new topics as represented by the socio-cognitive dynamics of these systems. We especially plan to investigate models developed in other domains, esp. biology (ecology, evolutionary theory) and cultural epidemiology [3, 5].

*** Skills required ***

This research is by nature multidisciplinary and a wide range of interest and skills would be appreciated. We will examine the following skills in the first instance :

- excellent programming skills
- interest for natural language processing and very large corpora
- interest for artificial intelligence, and more specifically machine learning
- a good command of French is necessary, as well as excellent writing skills

The candidate is expected to work at LaTTiCe and ISC-PIF, depending on the nature of the work. Both laboratories are located within Paris and are easily reachable using public transportation.

*** How to apply? ***

Send an email to Thierry Poibeau and Jean-Philippe Cointet with a CV and a cover letter as soon as possible and in any case before July, 17. Interviews will be held using Skype in July. The selection will be done before the end of July. The successful candidate is expected to begin his/her research in the beginning of October or November 2011.

*** Références ***

[1] D. Crandall, D. Cosley, D. Huttenlocher, J. M. Kleinberg, and S. Suri. Feedback effects between similarity and social influence in online communities. ACM New York, NY, USA, 2008.
[2] C. Jacquemin and P. Zweigenbaum. Traitement automatique des langues pour l’accès au contenu des documents. Le document en sciences du traitement de l’information, pages 71–109, 2000.
[3] J. Leskovec, L. Backstrom, and J. M. Kleinberg. Meme-tracking and the Dynamics of the News Cycle. Proceedings of The Fifteenth ACM SIGKDD International Conference on Knowledge Discovery and DataMining (KDD-09), 2009.
[4] C. Roth and J.-P. Cointet. Social and Semantic Coevolution in Knowledge Networks. Social Networks, special issue on the “Dynamics of Social Networks”, 2009.
[5] M. Simmons and L. Adamic. Memes Online : Extracted, Subtracted, Injected, and Recollected. ICWSM 2011, 2011.
_______________________________________________
UNSUBSCRIBE from this page: http://mailman.uib.no/options/corpora
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora