Other formats:
BibTeX
LaTeX
RIS
@article{1377613,
author = {Antol, Matej and Dohnal, Vlastislav},
article_number = {1},
doi = {http://dx.doi.org/10.15388/Informatica.2017.118},
keywords = {kNN query;approximate search;query popularity;index structure;metric space},
language = {eng},
issn = {0868-4952},
journal = {Informatica},
title = {Popularity-Based Ranking for Fast Approximate kNN Search},
url = {https://informatica.vu.lt/journal/INFORMATICA/article/838/info},
volume = {28},
year = {2017}
}
TY - JOUR
ID - 1377613
AU - Antol, Matej - Dohnal, Vlastislav
PY - 2017
TI - Popularity-Based Ranking for Fast Approximate kNN Search
JF - Informatica
VL - 28
IS - 1
SP - 1-21
EP - 1-21
PB - Lithuanian Academy of Sciences
SN - 08684952
KW - kNN query;approximate search;query popularity;index structure;metric space
UR - https://informatica.vu.lt/journal/INFORMATICA/article/838/info
L2 - https://informatica.vu.lt/journal/INFORMATICA/article/838/info
N2 - Similarity searching has become widely available in many on-line archives of multimedia data. Users accessing such systems look for data items similar to their specific query object and typically refine results by re-running the search with a query from the results. We study this issue and propose a mechanism of approximate kNN query evaluation that incorporates statistics of accessing index data partitions. Apart from the distance between database objects, it also considers the prior query answers to prioritize index partitions containing frequently retrieved data, so evaluating repetitive similar queries more efficiently. We verify this concept in a number of experiments.
ER -
ANTOL, Matej and Vlastislav DOHNAL. Popularity-Based Ranking for Fast Approximate kNN Search. \textit{Informatica}. Lithuanian Academy of Sciences, 2017, vol.~28, No~1, p.~1-21. ISSN~0868-4952. Available from: https://dx.doi.org/10.15388/Informatica.2017.118.
|
