Abstract
Though the emerging live streaming industry has attracted growing attention, the dominant yanzhi category where streamers mostly interact with the audience through amateur talent shows and casual chats has not been widely investigated. To decode the mechanism behind the popularity of yanzhi streamers, this study draws on Dual Coding Theory (DCT) to posit that age estimated from a streamer’s face and voice can influence the level of viewer engagement. To validate our hypothesized relationships, 274 one-minute video records ahead of a viewer commenting or/and gifting were collected and analyzed via deep learning algorithms. Analytical results attest to the negative effects of both facial and vocal age on viewer engagement, while their interaction has a positive impact on viewer engagement.
| Original language | English |
|---|---|
| Pages (from-to) | 435-447 |
| Number of pages | 13 |
| Journal | Journal of Management Analytics |
| Volume | 9 |
| Issue number | 4 |
| DOIs | |
| Publication status | Published - 2022 |
Keywords
- age
- deep learning
- dual coding theory
- live streaming
- viewer engagement
ASJC Scopus subject areas
- Statistics and Probability
- Business, Management and Accounting (miscellaneous)
- Statistics, Probability and Uncertainty