Formação de grupos em MOOCs utilizando Particle Swarm Optimization
Nenhuma Miniatura disponível
Data
2016-02-26
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal de Goiás
Resumo
The MassiveOpenOnlineCourses(MOOCs)areonlinecourseswithopenenrollment
that involvingahugeamountofstudentsfromdifferentlocations,withdifferentback-
grounds andinterests.Thelargenumberofstudentsimpliesahugeandunmanageable
number ofinteractions.Thisfact,alongwiththedifferentinterestsofstudents,resulting
in low-qualityinteractions.Duetothelargenumberofstudents,alsobecomesunviable
composition manuallylearninggroups.DuetothesecharacteristicspresentinMOOCs,
a methodforforminggroupswasdevelopedinthiswork,asanattempttoattendthedi-
chotomy existsbetweenthecollective,whichinvolvestheformationofanonlinelearning
community onamassivescale,andindividual,withdifferentinterests,priorknowledge
and expectationsanddifferentleadershipprofiles.Fortheformationofgroups,anadapta-
tion ofParticleSwarmOptimizationalgorithmwasproposedbasedonthreecriteria,kno-
wledge level,interestsandleadershipprofiles,formingthengroupswithdifferentlevels
of knowledge,similarinterestsanddistributedleadership,providingbetterinteractionand
knowledgeconstruction.Werecreatedtwovariationsoftheproblem,withfivestudents
and theothersix.Basedoncomputationaltests,thealgorithmdemonstratedthatableto
attend thegroupingcriteriainasatisfactorycomputingtimeandismoreefficientthanthe
model randomgroupsformation.Thetestsalsodemonstratedthatthealgorithmisrobust
taking intoaccountthevariousdatasetsanditerationsvariations.Toevaluatethequality
of interactionsandknowledgebuildingingroupsformedbythemethod,Acasestudy
wasconducted;andfortheanalysisofthecollecteddiscourses,itwastakenasthebasis
twomodelsofdiscourseanalysisfoundintheliterature.Theresultsofthecasestudy
demonstrated thatthegroupsformedbytheproposedmethodachievedthebestresultsin
the interactionsandknowledgeconstruction,whencomparedwithgroupsthatdonotuse
it.
Descrição
Palavras-chave
MOOC , Cursos online , Massivos abertos , CSCL , Aprendizagem colaborativa , Aprendizagem a distância , PSO , Otimização porenxamedepartículas , Análise combinatória , Composição de grupos , Formaçãode grupos , MOOC , Massive open online courses , CSCL , Colaborative learning , Distance learning , PSO , Particle swarm optimization , Combinatorial analysis , Group composition , Group formation
Citação
ULLMANN, M. R. D. Formação de grupos em MOOCs utilizando Particle Swarm Optimization. 2016. 99 f. Dissertação (Mestrado em Ciência da Computação) - Universidade Federal de Goiás, Goiânia, 2016.