Accelerating Metric Filtering by Improving Bounds on Estimated
Distances

D 2020

Accelerating Metric Filtering by Improving Bounds on Estimated Distances

MÍČ, Vladimír a Pavel ZEZULA

Základní údaje

Originální název

Accelerating Metric Filtering by Improving Bounds on Estimated Distances

Autoři

MÍČ, Vladimír a Pavel ZEZULA

Vydání

Cham, Similarity Search and Applications: 13th International Conference, SISAP 2020, Copenhagen, Denmark, September 30 - October 2, 2020, Proceedings, od s. 3-17, 15 s. 2020

Nakladatel

Springer

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Švýcarsko

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

tištěná verze "print"

Odkazy

URL

Impakt faktor

Impact factor: 0.402 v roce 2005

Označené pro přenos do RIV

Ano

Kód RIV

RIV/00216224:14330/20:00116699

Organizační jednotka

Fakulta informatiky

ISBN

978-3-030-60935-1

ISSN

Klíčová slova anglicky

Metric space;Similarity search;Triangle inequality;Metric filtering;Estimating unknown distance

Štítky

best, DISA, firank_B

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 10. 5. 2021 06:02, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

Filtering is a fundamental strategy of metric similarity indexes to minimise the number of computed distances. Given a triple of objects for which distances of two pairs are known, the lower and upper bounds on the third distance can be set as the difference and the sum of these two already known distances, due to the triangle inequality rule of the metric space. For efficiency reasons, the tightness of bounds is crucial, but as angles within triangles of distances can be arbitrary, the worst case with zero and straight angles must also be considered for correctness. However, in data of real-life applications, the distribution of possible angles is skewed and extremes are very unlikely to occur. In this paper, we enhance the existing definition of bounds on the unknown distance with information about possible angles within triangles. We show that two lower bounds and one upper bound on each distance exist in case of limited angles. We analyse their filtering power and confirm high improvements of efficiency by experiments on several real-life datasets.

Návaznosti

EF16_019/0000822, projekt VaV

Název: Centrum excelence pro kyberkriminalitu, kyberbezpečnost a ochranu kritických informačních infrastruktur

Přiložené soubory

2020-SISAP-Accelerating-metric-filtering-video.mp4

2020-SISAP-Accelerating-metric-filtering-poster.pdf

2020-SISAP-Accelerating-metric-filtering.pdf

2020-SISAP-Accelerating-metric-filtering-presentation.pdf

Citovat

MÍČ, Vladimír a Pavel ZEZULA. Accelerating Metric Filtering by Improving Bounds on Estimated Distances. In Shin'ichi Satoh, Lucia Vadicamo, Arthur Zimek, Fabio Carrara, Ilaria Bartolini, Martin Aumüller, Björn Þór JónssonRasmus Pagh. Similarity Search and Applications: 13th International Conference, SISAP 2020, Copenhagen, Denmark, September 30 - October 2, 2020, Proceedings. Cham: Springer, 2020, s. 3-17. ISBN 978-3-030-60935-1. Dostupné z: https://doi.org/10.1007/978-3-030-60936-8_1.

@inproceedings{1686161,
   author = {Míč, Vladimír and Zezula, Pavel},
   address = {Cham},
   booktitle = {Similarity Search and Applications: 13th International Conference, SISAP 2020, Copenhagen, Denmark, September 30 - October 2, 2020, Proceedings},
   doi = {https://doi.org/10.1007/978-3-030-60936-8_1},
   editor = {Shin'ichi Satoh, Lucia Vadicamo, Arthur Zimek, Fabio Carrara, Ilaria Bartolini, Martin Aumüller, Björn Þór JónssonRasmus Pagh},
   keywords = {Metric space;Similarity search;Triangle inequality;Metric filtering;Estimating unknown distance},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Cham},
   isbn = {978-3-030-60935-1},
   pages = {3-17},
   publisher = {Springer},
   title = {Accelerating Metric Filtering by Improving Bounds on Estimated Distances},
   url = {https://link.springer.com/chapter/10.1007/978-3-030-60936-8_1},
   year = {2020}
}

TY  - CONF
ID  - 1686161
AU  - Míč, Vladimír - Zezula, Pavel
PY  - 2020
TI  - Accelerating Metric Filtering by Improving Bounds on Estimated Distances
PB  - Springer
CY  - Cham
SN  - 9783030609351
KW  - Metric space;Similarity search;Triangle inequality;Metric filtering;Estimating unknown distance
UR  - https://link.springer.com/chapter/10.1007/978-3-030-60936-8_1
L2  - https://link.springer.com/chapter/10.1007/978-3-030-60936-8_1
N2  - Filtering is a fundamental strategy of metric similarity indexes to minimise the number of computed distances. Given a triple of objects for which distances of two pairs are known, the lower and upper bounds on the third distance can be set as the difference and the sum of these two already known distances, due to the triangle inequality rule of the metric space. For efficiency reasons, the tightness of bounds is crucial, but as angles within triangles of distances can be arbitrary, the worst case with zero and straight angles must also be considered for correctness. However, in data of real-life applications, the distribution of possible angles is skewed and extremes are very unlikely to occur. In this paper, we enhance the existing definition of bounds on the unknown distance with information about possible angles within triangles. We show that two lower bounds and one upper bound on each distance exist in case of limited angles. We analyse their filtering power and confirm high improvements of efficiency by experiments on several real-life datasets.
ER  -

MÍČ, Vladimír a Pavel ZEZULA. Accelerating Metric Filtering by Improving Bounds on Estimated Distances. In Shin'ichi Satoh, Lucia Vadicamo, Arthur Zimek, Fabio Carrara, Ilaria Bartolini, Martin Aumüller, Björn Þór JónssonRasmus Pagh. \textit{Similarity Search and Applications: 13th International Conference, SISAP 2020, Copenhagen, Denmark, September 30 - October 2, 2020, Proceedings}. Cham: Springer, 2020, s.~3-17. ISBN~978-3-030-60935-1. Dostupné z: https://doi.org/10.1007/978-3-030-60936-8\_{}1.

Přehled o publikaci