Who is Earning? Understanding and Modeling the Virtual Gifts Behavior of Users in Live Streaming Economy

Wei Tu,Chen Yan,Yiping Yan,Xu Ding,Lifeng Sun

DOI: https://doi.org/10.1109/mipr.2018.00028

2018-01-01

Abstract:Recently, live streaming is gaining increasing en- thusiasm, like Twitch, Inke and YouTube. Online streamers can interact with viewers in real time in live platforms with audio, video and text. The virtual gift is one of the most important business models for these live streaming platforms. The viewers can purchase and send virtual gifts to the streamers during the live process. These behaviors have great market potential and economic value, but little previous research paid attention about this. In this paper, we measured the basic information and economic behavior of users on a large representative live streaming platform (Inke1). To be specific, we discovered 20% of viewers contributed 97% of all gifts, and 77% of viewers only present for top three streamers. Especially, we also found that most gifting behaviors broke out in short periods when barrage increased, and gifting behaviors could be inspired by others due to competitiveness. Moreover, we investigated the correlation of various factors, proposed a decision tree based prediction model to infer the future revenue of the streamers. The result will bring benefit for both industrial entrepreneurs (for improving the design of the system operating mechanism of live platforms) and researchers (for interesting problems in personalized recommendation, etc.).