Rank Aggregation of Candidate Sets for Efficient Similarity
Search

D 2014

Rank Aggregation of Candidate Sets for Efficient Similarity Search

NOVÁK, David and Pavel ZEZULA

Basic information

Original name

Rank Aggregation of Candidate Sets for Efficient Similarity Search

Authors

NOVÁK, David (203 Czech Republic, guarantor, belonging to the institution) and Pavel ZEZULA (203 Czech Republic, belonging to the institution)

Edition

Haidelberg, 25th International Conference on Database and Expert Systems Applications (DEXA 2014 ), p. 42-58, 17 pp. 2014

Publisher

Springer International Publishing Switzerland

Other information

Language

English

Type of outcome

Stať ve sborníku

Field of Study

20201 Electrical and electronic engineering

Country of publisher

Switzerland

Confidentiality degree

není předmětem státního či obchodního tajemství

Publication form

printed version "print"

Impact factor

Impact factor: 0.402 in 2005

RIV identification code

RIV/00216224:14330/14:00073743

Organization unit

Faculty of Informatics

ISBN

978-3-319-10084-5

ISSN

DOI

http://dx.doi.org/10.1007/978-3-319-10085-2_4

Keywords in English

Similarity Search; Metric Space; Approximation; Scalability

Abstract

V originále

Many current applications need to organize data with respect to mutual similarity between data objects. Generic similarity retrieval in large data collections is a tough task that has been drawing researchers’ attention for two decades. A typical general strategy to retrieve the most similar objects to a given example is to access and then refine a candidate set of objects; the overall search costs (and search time) then typically correlate with the candidate set size. We propose a generic approach that combines several independent indexes by aggregating their candidate sets in such a way that the resulting candidate set can be one or two orders of magnitude smaller (while keeping the answer quality). This achievement comes at the expense of higher computational costs of the ranking algorithm but experiments on two real-life and one artificial datasets indicate that the overall gain can be significant.

Links

GBP103/12/G084, research and development project

Name: Centrum pro multi-modální interpretaci dat velkého rozsahu

Investor: Czech Science Foundation

Citovat

NOVÁK, David and Pavel ZEZULA. Rank Aggregation of Candidate Sets for Efficient Similarity Search. In 25th International Conference on Database and Expert Systems Applications (DEXA 2014 ). Haidelberg: Springer International Publishing Switzerland, 2014, p. 42-58. ISBN 978-3-319-10084-5. Available from: https://dx.doi.org/10.1007/978-3-319-10085-2_4.

@inproceedings{1192665,
   author = {Novák, David and Zezula, Pavel},
   address = {Haidelberg},
   booktitle = {25th International Conference on Database and Expert Systems Applications (DEXA 2014 )},
   doi = {http://dx.doi.org/10.1007/978-3-319-10085-2_4},
   keywords = {Similarity Search; Metric Space; Approximation; Scalability},
   howpublished = {tištěná verze "print"},
   language = {eng},
   location = {Haidelberg},
   isbn = {978-3-319-10084-5},
   pages = {42-58},
   publisher = {Springer International Publishing Switzerland},
   title = {Rank Aggregation of Candidate Sets for Efficient Similarity Search},
   year = {2014}
}

TY  - JOUR
ID  - 1192665
AU  - Novák, David - Zezula, Pavel
PY  - 2014
TI  - Rank Aggregation of Candidate Sets for Efficient Similarity Search
PB  - Springer International Publishing Switzerland
CY  - Haidelberg
SN  - 9783319100845
KW  - Similarity Search
KW  - Metric Space
KW  - Approximation
KW  - Scalability
N2  - Many current applications need to organize data with respect to mutual similarity between data objects. Generic similarity retrieval in large data collections is a tough task that has been drawing researchers’ attention for two decades. A typical general strategy to retrieve the most similar objects to a given example is to access and then refine a candidate set of objects; the overall search costs (and search time) then typically correlate with the candidate set size. We propose a generic approach that combines several independent indexes by aggregating their candidate sets in such a way that the resulting candidate set can be one or two orders of magnitude smaller (while keeping the answer quality). This achievement comes at the expense of higher computational costs of the ranking algorithm but experiments on two real-life and one artificial datasets indicate that the overall gain can be significant.
ER  -

NOVÁK, David and Pavel ZEZULA. Rank Aggregation of Candidate Sets for Efficient Similarity Search. In \textit{25th International Conference on Database and Expert Systems Applications (DEXA 2014 )}. Haidelberg: Springer International Publishing Switzerland, 2014, p.~42-58. ISBN~978-3-319-10084-5. Available from: https://dx.doi.org/10.1007/978-3-319-10085-2\_{}4.

Detailed Information on Publication Record