Journal of theoretical and applied electronic commerce research, Год журнала: 2025, Номер 20(2), С. 54 - 54
Опубликована: Март 24, 2025
The increasing popularity of short videos has presented sellers with fresh opportunities to craft video advertisements that incorporate diverse modal information, each modality potentially having a different influence on consumer engagement. Understanding which information is most important in attracting consumers can provide theoretical support researchers. However, the dimensionality multimodal features often higher than available data, posing specific difficulties data analysis. Therefore, designing analysis framework needed comprehensively extract and reduce advertisements, thus analyzing are more for In this study, we chose TikTok as research subject, employed deep learning machine techniques from encompassing visual, acoustic, title, speech text features. Subsequently, introduced method based mixed-regularization sparse representation select variables. Ultimately, utilized multiblock partial least squares regression regress selected variables alongside additional scalar calculate block importance. empirical results indicate visual key factors influencing engagement, providing subsequent offering practical insights marketers.
Язык: Английский