LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL

                       THE UNIVERSITY OF ULSTER
                            FACULTY OF ARTS 
                 SCHOOL OF CREATIVE ARTS & TECHNOLOGIES 
 
               IS PLEASED TO ANNOUNCE A SEMINAR ENTITLED:

                        `Modelling learner emotions 
                    in game-based learning environments'

                                  by

                            Dr. Karla Muñoz

                                 from
  
                   Computer Science Research Institute
                       University of Ulster, Magee
                            Northern Ireland

                       FRIDAY, 22nd February, 2013
                               1.00 P.M.
                   Foyle Arts Building, Magee Campus 
                         Room MQ-213, 1st Floor     


                                Abstract

Virtual   (VLE)  and  game-based   learning  (GBL)   environments  are
frequently employed in teaching and especially with the recent upsurge
in online learning. It is observed that the key to ensuring a positive
learning  experience  is to  personalise  learning  through e.g.   the
inclusion  of Intelligent  Tutoring Systems  (ITSs)  modelling learner
emotions.  Current  approaches to modelling  emotion in ITSs  have the
following limitations:  (1) they are shown to  be successful primarily
in   laboratory  settings,  (2)   they  frequently   employ  Cognitive
Psychology theories arising from contexts other than education and (3)
model  performance is  reported with  percentage values  which  can be
misleading.   We  report here  on  a  computational  model of  learner
emotions,  which can  be applied  to GBL  environments,  employing the
Control-Value  Theory of  achievement emotions  (Pekrun et  al.  2007)
explaining the  determination of  emotions in education  through Value
and   Control    appraisals.    Our   computational    model   entails
representation,  recognition  and  reasoning about  learner  emotions.
Representation  of  learner  achievement  emotions employs  a  dynamic
sequence  of Bayesian Networks  where Probabilistic  Relational Models
(PRMs)  facilitate defining  these  networks, and  the Necessary  Path
Condition (NPC) algorithm in combination with Pearson correlations and
the results  of Binary/Multinomial Logistic Regression  (BLR, MLR) are
employed in defining network structure.  Expectation-Maximisation (EM)
learning  facilitates  learning  network parameters.   Recognition  of
emotions is conducted through contextual variables (e.g.  mouse focus,
requests  for  help),  answers  to  questions  in-game  dialogues  and
Galvanic Skin Response (GSR) physiological signals.  Our computational
model is implemented in  PlayPhysics, an emotional GBL environment for
teaching  Physics  at  undergraduate  level  incorporating  the  first
computer  implementation  of  Control-Value  Theory.   PlayPhysics  is
implemented with  Java, the Unity Game Engine,  JavaScript, Hugin Lite
and  3D Studio Max.   PlayPhysics Space  Adventure task  scenarios and
curiosity, storytelling  and usability  features are discussed  and we
present results  from evaluation of  our emotional student  model with
cross-validation and Cohen's Kappa.  Future work includes applications
in  sentiment   analysis,  i.e.   the  analysis   of  multimodal  data
incorporating opinions, attitudes, emotions, moods and feelings.
 
**********************************************************************
Biography:

Dr. Karla Muñoz has recently  completed her Ph.D.  in Computer Science
at the University  of Ulster (Magee) under the  supervision of Dr. Tom
Lunney and  Prof.  Paul  Mc Kevitt.  She  is a Certified  Associate in
Project  Management  (CAPM) and  an  Associate  Fellow  of the  Higher
Education Academy (AFHEA).  She has received a PG. Cert.  (Researcher)
(2012) and M.Sc.  in Computing and Intelligent Systems (2008) from the
University of Ulster and a B.Sc. in Electronic System Engineering from
Tecnológico de Monterrey, Mexico City,  Mexico (2005). In 2008 she was
awarded a Vice Chancellor’s Research Scholarship (VCRS) and the 8over8
Prize   for  best   overall   performing  Master's   student  on   the
M.Sc. Computing with Specialism course at the University of Ulster. In
2007  she was  granted a  European  Union (EU)  scholarship for  Latin
America  (ALβAN)  in  order  to  pursue  her  M.Sc.   studies  at  the
University of Ulster.  She has worked as a researcher within industry,
such as  360 Production Ltd.,  Derry/London.  Dr. Muñoz  has published
research papers  in international journals,  conferences and workshops
and has represented the University of Ulster at SET for Britain, House
of Commons, London (2011).  Her research interests include e-learning,
Problem-Based Learning, User-Modelling,  Virtual & Game-Based Learning
Environments, Intelligent  Tutoring Systems, Serious  Games, Affective
Computing, Sentiment  Analysis and Artificial  Intelligence techniques
for reasoning and making decisions under uncertainty.

http://karlamunoz.net/index.html
http://www.linkedin.com/in/karlamunozesquivel
http://www.facebook.com/karla.c.esquivel

All Welcome !

LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL