Estimating the prevalence of rare events — theory and practice
The Unofficial Google Data Science Blog
AUGUST 27, 2019
Of course, any mistakes by the reviewers would propagate to the accuracy of the metrics, and the metrics calculation should take into account human errors. If we could separate bad videos from good videos perfectly, we could simply calculate the metrics directly without sampling. The missing verdicts create two problems.
98 
Let's personalize your content