Header

UZH-Logo

Maintenance Infos

Nearest Neighbour Join with Groups and Predicates


Cafagna, Francesco; Böhlen, Michael Hanspeter; Bracher, Annelies (2015). Nearest Neighbour Join with Groups and Predicates. In: Proceedings of the ACM Eighteenth International Workshop on Data Warehousing and OLAP, Melbourne, Australia, 19 October 2015 - 23 October 2015, 39-48.

Abstract

This paper proposes the nearest neighbor join, r x T [G, Θ] s, with similarity on T, and integrated support for grouping attributes G and selection predicates Θ. The corresponding valuation algorithm, roNNJ, is robust and does not suffer from redundant fetches and index false hits, which are major performance bottlenecks in nearest neighbour joins that do not support grouping attributes and selection predicates. Our solution does not compute redundant fetches since it accesses the fact table only once, and uses the groups of the outer relation to limit the fact table to its relevant portions. We experimentally evaluate our solution using a data warehouse that manages analyses of animal feeds, and the TPC-H.

Abstract

This paper proposes the nearest neighbor join, r x T [G, Θ] s, with similarity on T, and integrated support for grouping attributes G and selection predicates Θ. The corresponding valuation algorithm, roNNJ, is robust and does not suffer from redundant fetches and index false hits, which are major performance bottlenecks in nearest neighbour joins that do not support grouping attributes and selection predicates. Our solution does not compute redundant fetches since it accesses the fact table only once, and uses the groups of the outer relation to limit the fact table to its relevant portions. We experimentally evaluate our solution using a data warehouse that manages analyses of animal feeds, and the TPC-H.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Downloads

0 downloads since deposited on 17 Feb 2016
0 downloads since 12 months

Additional indexing

Item Type:Conference or Workshop Item (Paper), refereed, original work
Communities & Collections:03 Faculty of Economics > Department of Informatics
Dewey Decimal Classification:000 Computer science, knowledge & systems
Language:English
Event End Date:23 October 2015
Deposited On:17 Feb 2016 14:13
Last Modified:17 Dec 2018 08:17
Publisher:ACM
ISBN:978-1-4503-3785-4
OA Status:Closed
Free access at:Publisher DOI. An embargo period may apply.
Publisher DOI:https://doi.org/10.1145/2736277.2741118
Other Identification Number:merlin-id:12920

Download