NDLI: Optimale Dimensionswahl bei der Bearbeitung des Similarity Join

Content Provider	Springer Nature Link
Author	Böhm, Christian Kriegel, Hans Peter
Copyright Year	2002
Abstract	Der Similarity Join spielt zunehmend eine Rolle bei verschiedenen Anwendungen des Data Mining. Obwohl bereits mehrere Algorithmen zur Auswertung dieser Grundoperation in modernen Datenbankanwendungen vorgeschlagen wurden, und trotz einer eindeutigen CPU-Dominanz dieser Algorithmen, gibt es bislang kaum Ansätze, die sich mit dem CPU-Aspekt beschäftigen. Wir schlagen in diesem Beitrag ein allgemeines Prinzip zur Reduktion von Distanzberechnungen vor, das bei vielen Grundalgorithmen für den Similarity Join, z.B. dem R-Tree Similarity Join und seinen Varianten, dem $\varepsilon$ -kdB-Tree oder einem Spatial Hash-Verfahren eingesetzt werden kann. Unsere Lösung besteht aus einem Plane-Sweep-ähnlichen Verfahren, bei dem die optimale Sortierungsdimension gemäß einem Wahrscheinlichkeitsmodell ermittelt wird. In einer umfangreichen experimentellen Studie weisen wir die Überlegenheit unseres Verfahrens sowohl gegenüber verschiedenen Join-Basisverfahren ohne Plane-Sweep-Auswertung als auch gegenüber dem einfachen Plane-Sweep-ähnlichen Verfahren ohne Dimensionswahl nach. The similarity join plays an increasing role in various applications of data mining. Several algorithms for the computation of this important database primitive of modern applications have been proposed. Although these algorithms are clearly CPU bound, until now no solution concentrates on the CPU aspect. In this paper we propose a general technique for reducing distance calculations. Our technique can be applied on top of many basic algorithms for the similarity join, such as the R-tree similarity join and its variants, the $\varepsilon$ -kdB-tree, or some spatial hashing method. Our solution is a method which is similar to plane sweeping. The sweep dimension is selected according to a probability model. In an extensive experimental evaluation, we show the superiority of our approach with respect to different basic similarity join algorithms as well as with respect to a simple sweeping without selection of the optimal dimension.
Starting Page	68
Ending Page	76
Page Count	9
File Format	PDF
ISSN	01783564
Journal	Informatik - Forschung und Entwicklung
Volume Number	17
Issue Number	2
Language	German
Publisher	Springer-Verlag
Publisher Date	2002-07-01
Publisher Place	Berlin, Heidelberg
Access Restriction	Subscribed
Content Type	Text
Resource Type	Article
Subject	Computer Science

Sl.	Authority	Responsibilities	Communication Details
1	Ministry of Education (GoI), Department of Higher Education	Sanctioning Authority	https://www.education.gov.in/ict-initiatives
2	Indian Institute of Technology Kharagpur	Host Institute of the Project: The host institute of the project is responsible for providing infrastructure support and hosting the project	https://www.iitkgp.ac.in
3	National Digital Library of India Office, Indian Institute of Technology Kharagpur	The administrative and infrastructural headquarters of the project	Dr. B. Sutradhar bsutra@ndl.gov.in
4	Project PI / Joint PI	Principal Investigator and Joint Principal Investigators of the project	Dr. B. Sutradhar bsutra@ndl.gov.in Prof. Saswat Chakrabarti will be added soon
5	Website/Portal (Helpdesk)	Queries regarding NDLI and its services	support@ndl.gov.in
6	Contents and Copyright Issues	Queries related to content curation and copyright issues	content@ndl.gov.in
7	National Digital Library of India Club (NDLI Club)	Queries related to NDLI Club formation, support, user awareness program, seminar/symposium, collaboration, social media, promotion, and outreach	clubsupport@ndl.gov.in
8	Digital Preservation Centre (DPC)	Assistance with digitizing and archiving copyright-free printed books	dpc@ndl.gov.in
9	IDR Setup or Support	Queries related to establishment and support of Institutional Digital Repository (IDR) and IDR workshops	idr@ndl.gov.in