Paper Info
Title
Seeker: Topic-Aware Viewing Pattern Prediction in Crowdsourced Interactive Live Streaming
Abstract
Recently, Crowdsourced Interactive Live Streaming (CILS), such as Twitch.tv and Periscope, has emerged as one of the most popular streaming applications over the Internet. In such applications, a large number of geo-distributed users publish live sources to broadcast their game sessions, personal activities, and other events, while fellow viewers not only watch these live streams, but also contribute interactive messages to influence streaming content. Such explosively increasing popularity has posed significant challenges to predict viewing patterns using traditional time-series approaches, which lack the start/end knowledge of live streams and cannot capture the viewing burst very well. In this paper, we closely examine the characteristics of interactive messages in the real-world datasets, we find that the strong topic relevances exist in the viewers' discussions. Motivated by this observation, we design a crowdsourced framework Seeker to overcome aforementioned challenges. It explores the correlation between three viewing patterns (i.e., start/burst/end of live streams) and viewers' interactive messages (even before a live broadcast) through capturing the key topics. Our trace-driven evaluation and case study show the effectiveness of our solution, which can predict aforementioned patterns in advance and achieve much higher performance than the time-series approaches.
Year
DOI
Venue
2017
10.1145/3083165.3083179
NOSSDAV
Field
DocType
ISBN
Publication,Broadcasting,World Wide Web,Periscope,Computer science,Popularity,Live streaming,Multimedia,The Internet
Conference
978-1-4503-5003-7
Citations 
PageRank 
References 
4
0.52
5
Authors
5
Name
Order
Citations
PageRank
Cong Zhang1233.44
Jiangchuan Liu24340310.86
Ming Ma38715.25
Lifeng Sun496798.43
Bo Li5807.43