Predicting video virality and viewer engagement: a biometric data and machine learning approach

Dinko Bačić and Curt Gilstrap

Behaviour and Information Technology, 2024, vol. 43, issue 12, 2854-2880

Abstract: Predicting video virality is the holy grail of marketing today. Previous video virality prediction research has relied upon two processes that may lead to incomplete data or incomplete analyses: non-subconscious data collection and self-reporting. This exploratory study evaluates the potential of using a physiological manifestation of emotions captured through facial expressions and skin conductance to predict video viewer engagement across the viewing experience. In the context of video virality and user-focused emotional response, an experiment with 64 subjects viewing 13 videos collected facial expression and galvanic skin response data during the entire viewing experience. XGBoost classifier was deployed using 42 collected features (physiological data and socio-behavioural responses). The selected classifier could predict user engagement with over 80% accuracy. In addition to socio-demographic data and behavioural features, the predictive model identified several facial expression-based features, including action units (mouth open, cheek raise, eye closure, lip raise, and smile), emotions (joy), engagement and positive valence along with head movement (attention) and arousal (GSR peaks) as the most impactful. This study confirms the predictive capability of physiological data and elevates the value and need to understand further the role of viewers’ physiological and subconscious responses to video content across the viewing experience.

Date: 2024
References: Add references at CitEc
Citations:

Downloads: (external link)
http://hdl.handle.net/10.1080/0144929X.2023.2260894 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:taf:tbitxx:v:43:y:2024:i:12:p:2854-2880

Ordering information: This journal article can be ordered from
http://www.tandfonline.com/pricing/journal/tbit20

DOI: 10.1080/0144929X.2023.2260894

Access Statistics for this article

Behaviour and Information Technology is currently edited by Dr Panos P Markopoulos

More articles in Behaviour and Information Technology from Taylor & Francis Journals
Bibliographic data for series maintained by Chris Longhurst ().