Abstract
There are several interesting applications which are based on sequentially sampling individuals from an underlying population. Examples include online cardinality estimation [1-4] where the goal is to approximate the total size of the population and algorithms are typically based on the counting of 'collisions', i.e., instances where the sampled individual was 'seen' before; community exploration [5-7] where the goal is to discover as many distinct entities of interest as possible; and community detection / clustering [8-10] where the underlying population is naturally divided into communities / clusters and the goal is to estimate the true clustering using samples from the population.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
