DMCA
Counting youtube videos via random prefix sampling (2011)
| Venue: | In SIGCOMM |
| Citations: | 10 - 1 self |
BibTeX
@INPROCEEDINGS{Zhou11countingyoutube,
author = {Jia Zhou and Yanhua Li and Vijay Kumar Adhikari and Zhi-li Zhang},
title = {Counting youtube videos via random prefix sampling},
booktitle = {In SIGCOMM},
year = {2011},
pages = {371--380},
publisher = {ACM}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Leveraging the characteristics of YouTube video id space and ex-ploiting a unique property of YouTube search API, in this paper we develop a random prefix sampling method to estimate the total number of videos hosted by YouTube. Through theoretical model-ing and analysis, we demonstrate that the estimator based on this method is unbiased, and provide bounds on its variance and confi-dence interval. These bounds enable us to judiciously select sam-ple sizes to control estimation errors. We evaluate our sampling method and validate the sampling results using two distinct collec-tions of YouTube video id’s (namely, treating each collection as if it were the “true ” collection of YouTube videos). We then apply our sampling method to the live YouTube system, and estimate that there are a total of roughly 500 millions YouTube videos by May, 2011. Finally, using an unbiased collection of YouTube videos sampled by our method, we show that YouTube video view count statistics collected by prior methods (e.g., through crawling of re-lated video links) are highly skewed, significantly under-estimating the number of videos with very small view counts (< 1000) ; we al-so shed lights on the bounds for the total storage YouTube must have and the network capacity needed to delivery YouTube videos.
Keyphrases
random prefix youtube video youtube video id space live youtube system shed light estimation error small view count network capacity million youtube video select sam-ple size re-lated video link unbiased collection theoretical model-ing true collection delivery youtube video total number distinct collec-tions youtube search api prior method unique property confi-dence interval total storage youtube youtube video id sampling result view count statistic
