会话(web分析)
直播流媒体
价值(数学)
计算机科学
代理(统计)
社会化媒体
背景(考古学)
利润(经济学)
掉期(金融)
多媒体
广告
万维网
业务
古生物学
财务
机器学习
经济
生物
微观经济学
作者
Han Zhang,Yang Yang,Jichang Zhao
标识
DOI:10.1016/j.chb.2022.107467
摘要
Game live streaming has become increasingly popular and web streamer is widely recognized as a real career. A streamer is anticipated to entertain viewers, build relationships with them and solicit high-value gifts from viewers for profit. We explore game live streaming from a business aspect and find session-level factors of gifting with a data-driven approach. Previous studies extensively investigated the financial aspects of game live streaming and the social motivations behind this newly-emerged social media, but mostly from the individual level or as case studies. This study uses real live stream chatting and gifting data of over 1000 h and proposes metrics to proxy in-session behaviors for quantitative analysis of mass scale. We propose to use viewer chatting as a session-level measurement to better understand the motivations of high-value gifting. We find that the ratio of game-irrelevant chats (GIC) positively promotes high-value gifting, while small burst in game-irrelevant chats (spike) and frequent exchange of dominant topics (swap) negatively influence high-value gifting. Two sources of interactions, streamer and other viewers, elicit a framework to explain how in-session performance affects the high-value gifting. Specifically, parasocial relationship (PSR) with the streamer, and group identification (GI) with peer viewers are identified as mediators of the aforementioned three variables, and the two also positively influence high-value gifting. This research contributes to existing theories of gifting in live streaming context, and provides practical insights for live streaming platforms and streamers.
科研通智能强力驱动
Strongly Powered by AbleSci AI