Building an general purpose approximation mechanism in Hadoop, support approximations on any subset of the original data set.

Here is a link to the research paper of this project published on VLDB 2017

Acknowledgement

This work is supported in part by the US National Science Foundation Grant CCF-1337244, CCF-1527249, and National Science Foundation Early Career Award 0953946.

This work is also supported in part by the National Science Foundation under awards CNS-1042537 and CNS-1042543 (PRObE). http://www.nmc-probe.org/