Job: Post-doctoral position, Visual Analytics, University of Montpellier 2, France

Thierry Hamon hamon at LIMSI.FR
Sun Apr 27 07:11:12 UTC 2014

Date: Sat, 26 Apr 2014 15:19:59 +0200
From: Mathieu Roche <mathieu.roche at>
Message-Id: <168677E3-951A-4CAF-98FF-6735839DF21E at>

Post-doctoral position in Visual Analytics 

Location: LIRMM Lab (,
Montpellier (, France

Title: Visual Analytics for heterogeneous text streams.

The heterogeneity of data is an important issue in Big Data area. Data
is not only large in volume and produced at a high speed (velocity), but
also holds many kinds of input (technical heterogeneity), structures
(data model heterogeneity), and meanings (semantic heterogeneity). We
would like to recruit a postdoctoral researcher for one year to address
this specific issue from a visual analytics approach.

A lot of documents (web pages, scientific publications, reports, and so
forth) contain much useful information. Mining heterogeneous data,
according to their structure and to their content, becomes a major issue
in data mining area. A key problem consists of sharing these various
data and/or information and integrating them in order to discover new
knowledge. In addition, microbloggings contain crucial information to
take into account in a global system that mines heterogeneous data. For
instance, people participating in on-line forums, microblogging or
discussing on social networks leave behind them digital traces of
information on a variety of topics. The analysis of individual messages
and their aggregation represent a considerable challenge for currently
existing methods, as user-written texts present a special type of stream
setting. In this project we plan to investigate the epidemiology issue
of farmed animals in collaboration with UMR CMAEE. The aim is to detect
weak signals concerning the beginning of epidemics (e.g., African swine
fever, foot and mouth disease, bluetongue, avian influenza).

Visual exploration of textual data is an active area of research. Most
of the methods proposed deal with static texts like discourses, books
or, more generally, string data. Most of these methods require
well-formatted data and are not adapted to streams and/or heterogeneous
data. The candidate will be in charge of discovering efficient text
mining techniques for extracting structured data, and designing visual
interfaces to interact with these structures and explore the
data. He/She will process following the steps of the Munzner’s nested
model for visualization design. (1) Domain problem characterization: the
candidate will learn about the tasks and the data of the target domain
(2) Data/operation abstraction design: he/she will design the text
mining techniques that transform the raw data into the data types that
visualization techniques can address. (3) Encoding/interaction technique
design: he/she will design the visual encodings and interactions. (4)
Algorithm Design: he/she will create algorithms to carry out the visual
encodings and interactions designs automatically. The candidate will
also be in charge of organizing the validation process included in the
Munzner’s model.

- A PhD degree in Computer Science on the domain of Information
  Visualization or Visual Analytics with interest in Text Mining.
- An excellent publication record, including papers in high-impact
  journals and conference proceedings.
- Strong experience in programming languages.
- Knowledge of visualization libraries (e.g. D3, GraphViz, ...) is an
- Must be proficient in English.

- The project will be formally jointly supervised by Dr. Arnaud
  Sallaberry at LIRMM ( and Dr. Mathieu
  Roche at TETIS (

Starting date:
- September/October 2014 (some flexibility is possible)

Conditions of employment:
- Net salary: ~ 2200 euros / month
- Interested candidates are requested to send an application by e-mail
  to Dr. Arnaud Sallaberry (arnaud.sallaberry at and Dr. Mathieu
  Roche (mathieu.roche at with the subject field: 'LABEX:
  Post-Doc visualization position'.
- The application should consist of a motivation letter and a curriculum
  vitae with a list of publications and description of any previous
  research. Furthermore, names and contact information for three
  references are required (with two letters of recommendation).
- Applications will be reviewed immediately and the review process will
  continue until the position is filled.

Message diffuse par la liste Langage Naturel <LN at>
Informations, abonnement :
English version       : 
Archives                 :

La liste LN est parrainee par l'ATALA (Association pour le Traitement
Automatique des Langues)
Information et adhesion  :

ATALA décline toute responsabilité concernant le contenu des
messages diffusés sur la liste LN

More information about the Ln mailing list