Popularity-Based Ranking for Fast Approximate kNN Search

This publication doesn't include Faculty of Arts. It includes Faculty of Informatics. Official publication website can be found on muni.cz.

Authors

ANTOL Matej DOHNAL Vlastislav

Type Article in Periodical
Magazine / Source Informatica
MU Faculty or unit

Faculty of Informatics

Citation
WWW https://www.mii.lt/informatica/htm/INFO1130.htm
Doi http://dx.doi.org/10.15388/Informatica.2017.118
Field Informatics
Keywords kNN query;approximate search;query popularity;index structure;metric space
Description 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.
Related projects: