Interactive Exploration of Subspace Clusters on Multicore Processors

The Hai Pham, Jesper Kristensen, Thai Son Mai, Ira Assent, Jon Jacobsen, Bay Vo, Anh Le

Research output: Chapter in Book/Report/Conference proceedingChapter

Abstract

The PreDeCon clustering algorithm finds arbitrarily shaped clusters in high-dimensional feature spaces, which remains an active research topic with many potential applications. However, it suffers from poor runtime performance, as well as a lack of user interaction. Our new method AnyPDC introduces a novel approach to cope with these problems by casting PreDeCon into an anytime algorithm. In this anytime scheme, it quickly produces an approximate result and iteratively refines it toward the result of PreDeCon at the end. AnyPDC not only significantly speeds up PreDeCon clustering but also allows users to interact with the algorithm during its execution. Moreover, by maintaining an underlying cluster structure consisting of so-called primitive clusters and by block processing of neighborhood queries, AnyPDC can be efficiently executed in parallel on shared memory architectures such as multi-core processors. Experiments on large real world datasets show that AnyPDC achieves high quality approximate results early on, leading to orders of magnitude speedup compared to PreDeCon. Moreover, while anytime techniques are usually slower than batch ones, the algorithmic solution in AnyPDC is actually faster than PreDeCon even if run to the end. AnyPDC also scales well with the number of threads on multi-cores CPUs.
Original languageEnglish
Title of host publicationTransactions on Large-Scale Data- and Knowledge-Centered Systems XXXIX
PublisherSpringer
Pages169-199
DOIs
Publication statusPublished - 23 Nov 2018
Externally publishedYes

Publication series

NameTransactions on Large-Scale Data- and Knowledge-Centered Systems
PublisherSpringer

Fingerprint Dive into the research topics of 'Interactive Exploration of Subspace Clusters on Multicore Processors'. Together they form a unique fingerprint.

Cite this