Proceedings. 20th International Conference on Data Engineering
DOI: 10.1109/icde.2004.1320011
|View full text |Cite
|
Sign up to set email alerts
|

Continuously maintaining quantile summaries of the most recent N elements over a data stream

Abstract: Statistics over the most recently observed data elements are often required in applications involving data streams, such as intrusion detection in network monitoring, stock price prediction in financial markets, web log mining for access prediction, and user click stream mining for personalization. Among various statistics, computing quantile summary is probably most challenging because of its complexity. In this paper, we study the problem of continuously maintaining quantile summary of the most recently obse… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

0
95
0

Publication Types

Select...
5
1
1

Relationship

0
7

Authors

Journals

citations
Cited by 59 publications
(95 citation statements)
references
References 24 publications
0
95
0
Order By: Relevance
“…In the sliding window model, where only the last W data elements are considered, the problem was first studied in [12] and improved in [1] to space O( 1 log 1 log W ).…”
Section: Related Workmentioning
confidence: 99%
“…In the sliding window model, where only the last W data elements are considered, the problem was first studied in [12] and improved in [1] to space O( 1 log 1 log W ).…”
Section: Related Workmentioning
confidence: 99%
“…Both deterministic and randomized methods are proposed. In our implementation, we adopt the method of computing approximate quantile summaries in a sliding window proposed in [41], which is based on the GK-algorithm [26] that finds the approximate quantile over a data steam. The algorithm can continuously output the -approximate quantiles in a sliding window with space cost of O( log 2 ω 2 ).…”
Section: Definition 11 ( -Approximate Quantile) Let O 1 ≺ · · · ≺ O ωmentioning
confidence: 99%
“…Computing -Approximate quantiles in data streams is well studied [25,26,41,43]. Both deterministic and randomized methods are proposed.…”
Section: Definition 11 ( -Approximate Quantile) Let O 1 ≺ · · · ≺ O ωmentioning
confidence: 99%
“…The algorithm uses O(log 2 |U | log (log(|U |)/δ)/ 2 ) space, given the value domain U of data and the algorithm estimates -approximate quantiles with confidence 1 − δ. The only work we know for estimating quantiles on sliding windows is [6]. In this paper, the authors gave a solution for estimating quantiles on a sliding window containing a fixed number of data items.…”
Section: Introductionmentioning
confidence: 99%
“…Firstly, the number of data items in the window is unknown and changes over time. The technique in [6] does not work without an apriori to the window size. Secondly, constraints may have various of forms and a practical query processing should be able to handle different window queries in a uniform way rather than hacking from one specific solution to another.…”
Section: Introductionmentioning
confidence: 99%