<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-forward-container"><br>
<div style="text-align: center;" class=""><b class=""><font
class="" size="4">PhD fellowship</font></b></div>
<div style="text-align: center;" class=""><b class=""><br class="">
</b></div>
<div style="text-align: center;" class=""><i class=""><font
class="" size="4">Towards a Computational Model of Social
Skills for Human-Computer Interaction</font></i></div>
<div style="text-align: center;" class=""><i class=""><font
class="" size="4"><br class="">
</font></i></div>
<div style="text-align: center;" class="">Aix-Marseille Université
& CNRS</div>
<div class=""><br class="">
</div>
<div class="">>>> Deadline for application: <b class="">April
9th, 2018</b> <br class="">
</div>
<div class=""><br class="">
</div>
The goal of this project is to develop a computational model of
social skills for multimodal interaction systems. The model will
focus more precisely on a specific context, task oriented
dialogues, in which all semantic and pragmatic aspects are
controlled. It will be built by means on different machine
learning methodologies applied to the analysis of natural corpora.
This work will be part of an ongoing project (<a
href="http://www.lpl-aix.fr/%7Eacorformed/" class=""
moz-do-not-send="true">http://www.lpl-aix.fr/~acorformed/</a>),
aiming at developing a virtual reality system for training doctors
to break bad news, with an embodied conversational agent playing
the role of a virtual patient.
<div class="">
<div class=""><br class="">
</div>
<div class="">The PhD is organized around 4 main tasks: <br
class="">
<ul class="">
<li class="">Describing the social skills relevant to the
use case, with a specific focus on empathy and persuasion,
through a corpus analysis. This task is done in
collaboration with doctors involved in such training
activities.</li>
<li class="">Modeling the social skills, by means of machine
learning techniques</li>
<li class="">Implementing the social skills in the
communication environment</li>
<li class="">Developing a tool for the automatic evaluation
of doctor's skills, adapted to the training goals</li>
</ul>
</div>
<div class=""><br class="webkit-block-placeholder">
</div>
<div class="">The PhD candidate should have a master's degree
completed in one of the fields below:<br class="">
<ul class="">
<li class="">Computer science</li>
<li class="">Artificial Intelligence</li>
<li class="">Natural language processing</li>
<li class="">Applied mathematics</li>
</ul>
The candidate should have a strong background in machine
learning and modeling methods. Some complementary previous
experience would be appreciated in the following topics:<br
class="">
<ul class="">
<li class="">Human-computer interaction</li>
<li class="">Multimodal data processing</li>
<li class="">Data acquisition</li>
<li class="">Corpus-based studies</li>
<li class="">Affective computing</li>
<li class="">Conversational agents</li>
</ul>
<br class="">
This fellowship is part of the Marie Skłodowska-Curie COFUND
action. It is a three-year work contract, with a monthly net
salary of approximately 1625€/month, a 500€ travel allowance
per year between Marseille and the place of origin, a
financial support for international research training and
conferences participations plus a contribution to the research
costs of up to 12.600€. </div>
<div class=""><br class="">
</div>
<div class="">More information here: <a
href="https://doc2amu.univ-amu.fr/en/towards-a-computational-model-of-social-skills-for-human-computer-interaction-so-human"
class="" moz-do-not-send="true">https://doc2amu.univ-amu.fr/en/towards-a-computational-model-of-social-skills-for-human-computer-interaction-so-human</a> <br
class="">
</div>
<div class=""><br class="webkit-block-placeholder">
</div>
<div class="">
<div class="">For any question, contact Philippe Blache <<font
color="#000000"><a
href="mailto:philippe.blache@univ-amu.fr" class=""
moz-do-not-send="true">philippe.blache@univ-amu.fr</a></font>>
or Magalie Ochs <a class="moz-txt-link-rfc2396E" href="mailto:magalie.ochs@lis-lab.fr"><magalie.ochs@lis-lab.fr ></a></div>
</div>
<div class=""><br class="">
</div>
<br class="">
</div>
</div>
</body>
</html>