Non-Linear Consumption of Videos Using a Sequence of Personalized Multimodal Fragments

Verma, Gaurav; Nalamada, Trikay; Harpavat, Keerti; Goel, Pranav; Mishra, Aman Kumar; Srinivasan, Balaji Vasan

doi:10.1145/3397481.3450672

Cited by 10 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous studies have suggested that users are sensitive to content type only when video length varies to a substantial degree (Guo et al. , 2014; Verma et al. , 2021).…”

Section: Discussionmentioning

confidence: 99%

“…First, the limited variation in video length that characterizes TikTok meant that the study scope precluded consideration of how video length moderates the relationship between algorithm-related factors and flow experience. Previous studies have suggested that users are sensitive to content type only when video length varies to a substantial degree (Guo et al, 2014;Verma et al, 2021). Therefore, there are limited insights regarding whether the recommendation algorithm functions similarly in the long-video context.…”

Section: Limitations and Recommendations For Future Researchmentioning

confidence: 99%

See 1 more Smart Citation

How TikTok leads users to flow experience: investigating the effects of technology affordances with user experience level and video length as moderators

Zhao

Wagner

2022

INTR

View full text Add to dashboard Cite

PurposeAlthough the short-video-based application TikTok and its AI-enhanced technology have achieved enormous success and reshaped the user experience, few studies have focused on the user experience in the TikTok context. This study adopts a technology affordance theory lens to identify the main mechanisms contributing to the user experience in the short-video platform context while including user experience level and video length as moderating effects.Design/methodology/approachThis study collected 401 valid questionnaires from TikTok users and used the structural equation modeling approach to examine the proposed research model.FindingsThree technology affordances (perceived effortlessness, perceived recommendation accuracy, and perceived recommendation serendipity) contribute to TikTok users' optimal flow experience. Multi-group analysis results further indicate that individuals react differently to the same stimuli as their experience level increases. Finally, video length critically influences the impact of technology affordances on users' cognitive responses.Originality/valueAs a burgeoning industry, the mechanisms enabling short-video platforms to engage users remain unclear to practitioners and researchers. Thus, this study's technology affordance lens provides necessary insights into how TikTok's innovative and advanced technologies contribute to user flow experience from a context-dependent perspective. Furthermore, given that most existing studies have neglected possible variations in user preferences when investigating the effects of technology, this study enriches the existing literature by employing user experience level and video length as moderators.

show abstract

“…Previous studies have suggested that users are sensitive to content type only when video length varies to a substantial degree (Guo et al. , 2014; Verma et al. , 2021).…”

Section: Discussionmentioning

confidence: 99%

Section: Limitations and Recommendations For Future Researchmentioning

confidence: 99%

How TikTok leads users to flow experience: investigating the effects of technology affordances with user experience level and video length as moderators

Zhao

Wagner

2022

INTR

View full text Add to dashboard Cite

show abstract

“…Video augmentation methods encompass different types of visual feedback, including video transcripts, concept maps, knowledge graphs and hypergraphs [5,6,27,33,36] that are specifically addressed in our project. Data driven investigations have explored possibilities for video augmentation through personalized fragment navigation [41], enhanced exploration of e-learning contents [28,34], automatic display of relevant frames and visual summaries representing points with high learner activity [20]. The advancements in machine learning and natural language processing supported the production of interactive transcript of videos and keyword summary [20], and text summarization [7,20,23].…”

Section: Related Workmentioning

confidence: 99%

Video augmentation to support video-based learning

Torre

Galluccio

Coccoli

2022

Proceedings of the 2022 International Conference on Advanced Visual Interfaces

View full text Add to dashboard Cite

Multimedia content and video-based learning are expected to take a central role in the post-pandemic world. Thus, providing new advanced interfaces and services that further exploit their potential becomes of paramount importance. A challenging area deals with developing intelligent visual interfaces that integrate the knowledge extracted from multimedia materials into educational applications. In this respect, we designed a web-based video player that is aimed to support video consumption by exploiting the knowledge extracted from the video in terms of concepts explained in the video and prerequisite relations between them. This knowledge is used to augment the video lesson through visual feedback methods. Specifically, in this paper we investigate the use of two types of visual feedback, i.e. an augmented transcript and a dynamic concept map (map of concept's flow), to improve video comprehension in the first-watch learning context. Our preliminary findings suggest that both the methods help the learner to focus on the relevant concepts and their related contents. The augmented transcript has an higher impact on immediate comprehension compared to the map of concepts' flow, even though the latter is expected to be more powerful to support other tasks such as exploration and in-depth analysis of the concepts in the video.

show abstract

“…The way concept maps are integrated into hypervideo services is strictly correlated to the issue of information visualization and to the importance of the content presentation, in order to have a certain effectiveness, in the consolidated perspective that even in the educational field the learner can be considered as a prosumer [16]. Many projects based on a data driven approach have explored the different possibilities for improving the navigation experience, such as data-enhanced transcript search and keyword summary, automatic display of relevant frames, a visual summary representing points with high learner activity [17], non linear consumption of videos using personalized fragment navigation [32], exploration of e-learning contents via small screens [27]. The idea of using concept maps to support video navigation is already present in works at the beginning of the new millennium [13] but, at the best of the authors' knowledge, the novelty of this contribution can be identified in the possibility of automating the creation of concept maps, hopefully going in the direction of creating an on-the-fly service.…”

Section: Related Workmentioning

confidence: 99%