Relative Error Streaming Quantiles
Relative Error Streaming Quantiles
复制标题
相对误差流分位数
DOI:
10.1145/3617891
复制
发表时间:
2023
影响因子:
2.5
通讯作者:
Veselý, Pavel
中科院分区:
文献类型:
--
作者:
Cormode, Graham;Karnin, Zohar;Liberty, Edo;Thaler, Justin;Veselý, Pavel
Estimating ranks, quantiles, and distributions over streaming data is a central task in data analysis and monitoring. Given a stream ofnitems from a data universe equipped with a total order, the task is to compute a sketch (data structure) of size polylogarithmic inn. Given the sketch and a query itemy, one should be able to approximate its rank in the stream, i.e., the number of stream elements smaller than or equal toy.Most works to date focused on additive εnerror approximation, culminating in the KLL sketch that achieved optimal asymptotic behavior. This article investigatesmultiplicative(1± ε)-error approximations to the rank. Practical motivation for multiplicative error stems from demands to understand the tails of distributions, and hence for sketches to be more accurate near extreme values.The most space-efficient algorithms due to prior work store either O(log (ε2n)/ε2) orO(log3(εn)/ε) universe items. We present a randomized sketch storingO(log1.5(εn)/ε) items that can (1± ε)-approximate the rank of each universe item with high constant probability; this space bound is within anfactor of optimal. Our algorithm does not require prior knowledge of the stream length and is fully mergeable, rendering it suitable for parallel and distributed computing environments.
登录
查看更多内容
DOI:
10.1145/3375395.3387650
发表时间:
2020
期刊:
Proceedings of the 39th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems
影响因子:
--
作者:
Graham Cormode;P. Veselý
通讯作者:
P. Veselý
DOI:
10.1145/3471485.3471488
发表时间:
2020-03
期刊:
ACM SIGMOD Record
影响因子:
--
作者:
Omri Ben-Eliezer;Rajesh Jayaram;David P. Woodruff;E. Yogev
通讯作者:
Omri Ben-Eliezer;Rajesh Jayaram;David P. Woodruff;E. Yogev
影响因子:
1
作者:
David Felber;R. Ostrovsky
通讯作者:
R. Ostrovsky
DOI:
10.1145/1321440.1321601
发表时间:
2007
期刊:
Sensors (Basel, Switzerland)
影响因子:
--
作者:
Qi Zhang;Wei Wang
通讯作者:
Wei Wang
DOI:
--
发表时间:
2021
期刊:
Knowledge Discovery and Data Mining
影响因子:
--
作者:
Graham Cormode;Abhinav Mishra;Joseph Ross;P. Vesel'y
通讯作者:
P. Vesel'y