Development of an automatic attitude recognition system: a multimodal analysis of video blogs

Communicative content in human communication involves expressivity of socio-affectivestates. Research in Linguistics, Social Signal Processing and Affective Computing in particular, highlights the importance of affect, emotion and attitudes as sources of information forcommunicative content. Attitud...

Full description

Saved in:
Bibliographic Details
Main Author: Noor Alhusna Madzlan
Format: thesis
Language:eng
Published: 2017
Subjects:
Online Access:https://ir.upsi.edu.my/detailsg.php?det=5364
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Communicative content in human communication involves expressivity of socio-affectivestates. Research in Linguistics, Social Signal Processing and Affective Computing in particular, highlights the importance of affect, emotion and attitudes as sources of information forcommunicative content. Attitudes, considered as socio-affective states of speakers, areconveyed through a multitude of signals during communication. Understanding the expres sion ofattitudes of speakers is essential for establishing successful communication. Taking theempirical approach to studying attitude expressions, the main objective of this research isto contribute to the development of an automatic attitude classification system through afusion of multimodal signals expressed by speakers in video biogs. The present study describes a new communicative genre of self-expression through social media: video blogging, whichprovides opportunities for interlocutors to disseminate information through a myriad of multimodal characteristics. This study describes main features of this novel communica tion medium andfocuses attention to its possible exploitation as a rich source of information for humancommunication. The dissertation describes manual annotation of attitude expres sions from the vlogcorpus, multimodal feature analysis and processes for development of an automatic attitudeannotation system. An ontology of attitude annotation scheme for speech in video biogs iselaborated and five attitude labels are derived. Prosodic and visual fea tureextraction procedures are explained in detail. Discussion on processes of developing an automaticattitude classification model includes analysis of automatic prediction of attitude labelsusing prosodic and visual features through machine-learning methods. This study also elaboratesdetailed analysis of individual feature contributions and their predictive power tothe classification task