A Multimodal Approach to Confidence, Emotion, and Personality Assessment in Virtual Interviews
DOI :
https://doi.org/10.5269/bspm.82391Résumé
Therecruitmentlandscapeemphasizesprioritizingcandidatestechnicalskillsoverinterpersonal
skills suchasemotional intelligence, suitablebehaviour, communicationstyleandadaptability. Traditional
systemsanalysetextorskillbasedresponses,therebymissingouttoanalysebehaviourandrealtimeemotions
ininterview. Thiscreatesgapinanalysingthecandidatestruepersonalityprofileandstyleof interaction.
Therefore, there isnecessityofmultimodal systemtoevaluate ,whatcandidatespeak,howtheyspeakand
howtheypresentthemselvesduringinterview.Toaddressthis limitations, thisstudyproposesamultimodal
emotionrecognitionsystemwhichstrengthensforpersonalityassessmentindigitalinterviews.Thisframework
processesbyintegratinglinguisticcues fromtextual transcripts,paralinguisticcues fromspeechsignalsand
visualexpressionsfromvideoframes.Textbasedemotionfeaturesareextractedusingafinetunedtransformer
model,achieved43%accuracy. Audiobasedemotional cuesarerecognizedusingaspeechemotionpipeline,
achieved63%accuracy.Facialexpressionsareanalysedusingvisionbasedmodule.Earlyfusionstrategyhas
beenappliedto integratethesemultiplemodalitieswhichenables thesystemtogetconsolidatedemotional
statefromtheinterview.Thismultimodalarchitecturegivesagreatsupporttoevaluatecandidatecomplete
profile,whichmakes therecruitmentdecisionholistic. Thismakes theorganizations tohirethecandidates
whoarenotonlytechnicallystrongbutalsowithgoodinterpersonalskills.
Téléchargements
Publié
Numéro
Rubrique
Licence
© Boletim da Sociedade Paranaense de Matemática 2026
Cette œuvre est sous licence Creative Commons Attribution 4.0 International.
When the manuscript is accepted for publication, the authors agree automatically to transfer the copyright to the (SPM).
The journal utilize the Creative Common Attribution (CC-BY 4.0).
Comment citer
