2012 LinearSpaceDirectPatternSamplin: Difference between revisions

From GM-RKB
Jump to navigation Jump to search
(Importing text file)
 
m (Text replacement - " ↵↵" to " ")
 
(13 intermediate revisions by 2 users not shown)
Line 1: Line 1:
* ([[2012_LinearSpaceDirectPatternSamplin|Boley & al, 2012]]) ⇒ [[author::Mario Boley]], [[author::Sandy Moens]], and [[author::Thomas Gärtner]]. ([[year::2012]]). "Linear Space Direct Pattern Sampling Using Coupling from the Past." In: [[proceedings::Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining]] ([[conference::KDD-2012]]). ISBN:978-1-4503-1462-6 [http://dx.doi.org/10.1145/2339530.2339545 doi:10.1145/2339530.2339545]  
* ([[2012_LinearSpaceDirectPatternSamplin|Boley et al., 2012]]) [[author::Mario Boley]], [[author::Sandy Moens]], and [[author::Thomas Gärtner]]. ([[year::2012]]). “Linear Space Direct Pattern Sampling Using Coupling from the Past.In: [[proceedings::Proceedings of the 18th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining]] ([[conference::KDD-2012]]). ISBN:978-1-4503-1462-6 [http://dx.doi.org/10.1145/2339530.2339545 doi:10.1145/2339530.2339545]


<B>Subject Headings:</B>  
<B>Subject Headings:</B>


==Notes==
== Notes ==


==Cited By==
==Cited By==
Line 9: Line 9:
* http://dl.acm.org/citation.cfm?id=2339530.2339545&preflayout=flat#citedby
* http://dl.acm.org/citation.cfm?id=2339530.2339545&preflayout=flat#citedby


==Quotes==
== Quotes ==
===Author Keywords===
 
=== Author Keywords ===
* [[Cftp]]; [[data mining]]; [[frequent set]]s; [[local pattern]]s; [[sampling]]
* [[Cftp]]; [[data mining]]; [[frequent set]]s; [[local pattern]]s; [[sampling]]


===Abstract===
=== Abstract ===


[[This paper]] shows how [[coupling]] from the [[past (CFTP]]) can be used to [[avoid time]] and [[memory bottleneck]]s in [[direct local pattern sampling procedure]]s. </s>
[[This paper]] shows how [[coupling from the past (CFTP)]] can be used to [[avoid time]] and [[memory bottleneck]]s in [[direct local pattern sampling procedure]]s. </s>
Such procedures draw [[controlled amounts of suitably biased sample]]s directly from the [[pattern space]] of a given [[dataset]] in [[polynomial time]]. </s>
Such [[procedure]]s draw [[controlled amount]]s of suitably [[biased samples]] directly from the [[pattern space]] of a given [[dataset]] in [[polynomial time]]. </s>
Previous direct pattern [[sampling method]]s can produce [[pattern]]s in rapid succession after some initial preprocessing phase. </s>
Previous [[direct pattern sampling method]]s can produce [[pattern]]s in [[rapid succession]] after some initial [[preprocessing phase]]. </s>
This [[preprocessing phase]], however, turns out to be prohibitive in terms of time and memory for many [[dataset]]s. </s>
This [[preprocessing phase]], however, turns out to be prohibitive in terms of [[time]] and [[memory]] for many [[dataset]]s. </s>
[[We]] show how [[CFTP]] can be used to avoid any [[super-linear preprocessing]] and [[memory requirement]]s. </s>
[[We]] show how [[CFTP]] can be used to avoid any [[super-linear preprocessing]] and [[super-linear memory|memory requirement]]s. </s>
This allows to simulate more [[complex distribution]]s, which previously were intractable. </s>
This allows to [[simulate]] more [[complex distribution]]s, which previously were [[intractable]]. </s>
[[We]] show for a [[large number of public real-world dataset]]s that these [[new algorithm]]s are fast to [[execute]] and their [[pattern collection]]s outperform previous [[approach]]es both in [[unsupervised]] as well as [[supervised context]]s. </s>
[[We]] show for a [[large number]] of [[public real-world dataset]]s that these new [[algorithm]]s are fast to [[execute]] and their [[pattern collection]]s outperform previous [[approach]]es both in [[unsupervised]] as well as [[supervised context]]s. </s>


==References==
== References ==
{{#ifanon:|
{{#ifanon:|
* 1. Mohammad Al Hasan, Mohammed J. Zaki, Output Space Sampling for Graph Patterns, Proceedings of the VLDB Endowment, v.2 n.1, August 2009
* 1. Mohammad Al Hasan, Mohammed J. Zaki, Output Space Sampling for Graph Patterns, Proceedings of the VLDB Endowment, v.2 n.1, August 2009

Latest revision as of 19:33, 20 December 2023

Subject Headings:

Notes

Cited By

Quotes

Author Keywords

Abstract

This paper shows how coupling from the past (CFTP) can be used to avoid time and memory bottlenecks in direct local pattern sampling procedures. Such procedures draw controlled amounts of suitably biased samples directly from the pattern space of a given dataset in polynomial time. Previous direct pattern sampling methods can produce patterns in rapid succession after some initial preprocessing phase. This preprocessing phase, however, turns out to be prohibitive in terms of time and memory for many datasets. We show how CFTP can be used to avoid any super-linear preprocessing and memory requirements. This allows to simulate more complex distributions, which previously were intractable. We show for a large number of public real-world datasets that these new algorithms are fast to execute and their pattern collections outperform previous approaches both in unsupervised as well as supervised contexts.

References

;

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2012 LinearSpaceDirectPatternSamplinThomas Gärtner
Sandy Moens
Mario Boley
Linear Space Direct Pattern Sampling Using Coupling from the Past10.1145/2339530.23395452012