SC-LSH: Une Méthode d'Indexation pour une Recherche de Similarité Approximative dans l'Espace Multidimensionnel

Sanaa Chafik; Imane Daoudi; Mounim A. EL Yacoubi; Hamid El Ouardi

SC-LSH: Une Méthode d'Indexation pour une Recherche de Similarité Approximative dans l'Espace Multidimensionnel

Translated title of the contribution: SC-LSH. An indexing method for similarity search in multidimensional space approximate

Sanaa Chafik
, Imane Daoudi
, Mounim A. EL Yacoubi
, Hamid El Ouardi

Institut Mines-Télécom
Nancy Université
Hassan II Ain Chock University

Research output: Contribution to conference › Paper › peer-review

Abstract

Locality Sensitive Hashing (LSH) is one of the most promising techniques for solving nearest Neighbours search problem in high dimensional space. Euclidean LSH is the most popular variation of LSH that has been successfully applied in many multimedia applications. However, the Euclidean LSH presents limitations that affect search performances. The main limitation of the Euclidean LSH is the large memory consumption. In order to achieve a good accuracy, a large number of hash tables is required. This paper propose a new hashing algorithm to overcome the storage space problem, while keeping a good accuracy and better query time. The Experimental results on a real large-scale dataset show the interest of our approach..

Translated title of the contribution	SC-LSH. An indexing method for similarity search in multidimensional space approximate
Original language	French
Pages	303-318
Number of pages	16
Publication status	Published - 1 Jan 2015
Externally published	Yes
Event	Conference in Search Infomations and Applications, CORIA 2015 - 12th French Information Retrieval Conference - Paris, France Duration: 18 Mar 2015 → 20 Mar 2015

Conference

Conference	Conference in Search Infomations and Applications, CORIA 2015 - 12th French Information Retrieval Conference
Country/Territory	France
City	Paris
Period	18/03/15 → 20/03/15

Cite this

@conference{f9c76ca060344374bec7488209186dc1,

title = "SC-LSH: Une M{\'e}thode d'Indexation pour une Recherche de Similarit{\'e} Approximative dans l'Espace Multidimensionnel",

abstract = "Locality Sensitive Hashing (LSH) is one of the most promising techniques for solving nearest Neighbours search problem in high dimensional space. Euclidean LSH is the most popular variation of LSH that has been successfully applied in many multimedia applications. However, the Euclidean LSH presents limitations that affect search performances. The main limitation of the Euclidean LSH is the large memory consumption. In order to achieve a good accuracy, a large number of hash tables is required. This paper propose a new hashing algorithm to overcome the storage space problem, while keeping a good accuracy and better query time. The Experimental results on a real large-scale dataset show the interest of our approach..",

keywords = "Content based image retrieval (CBIR), Curse of dimensionality, LSH, Multidimensional indexing, Nearest neighbour search, Scalability",

author = "Sanaa Chafik and Imane Daoudi and \{EL Yacoubi\}, \{Mounim A.\} and \{El Ouardi\}, Hamid",

year = "2015",

month = jan,

day = "1",

language = "Fran{\c c}ais",

pages = "303--318",

note = "Conference in Search Infomations and Applications, CORIA 2015 - 12th French Information Retrieval Conference ; Conference date: 18-03-2015 Through 20-03-2015",

}

SC-LSH: Une Méthode d'Indexation pour une Recherche de Similarité Approximative dans l'Espace Multidimensionnel. / Chafik, Sanaa; Daoudi, Imane; EL Yacoubi, Mounim A. et al.
2015. 303-318 Paper presented at Conference in Search Infomations and Applications, CORIA 2015 - 12th French Information Retrieval Conference, Paris, France.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - SC-LSH

T2 - Conference in Search Infomations and Applications, CORIA 2015 - 12th French Information Retrieval Conference

AU - Chafik, Sanaa

AU - Daoudi, Imane

AU - EL Yacoubi, Mounim A.

AU - El Ouardi, Hamid

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Locality Sensitive Hashing (LSH) is one of the most promising techniques for solving nearest Neighbours search problem in high dimensional space. Euclidean LSH is the most popular variation of LSH that has been successfully applied in many multimedia applications. However, the Euclidean LSH presents limitations that affect search performances. The main limitation of the Euclidean LSH is the large memory consumption. In order to achieve a good accuracy, a large number of hash tables is required. This paper propose a new hashing algorithm to overcome the storage space problem, while keeping a good accuracy and better query time. The Experimental results on a real large-scale dataset show the interest of our approach..

AB - Locality Sensitive Hashing (LSH) is one of the most promising techniques for solving nearest Neighbours search problem in high dimensional space. Euclidean LSH is the most popular variation of LSH that has been successfully applied in many multimedia applications. However, the Euclidean LSH presents limitations that affect search performances. The main limitation of the Euclidean LSH is the large memory consumption. In order to achieve a good accuracy, a large number of hash tables is required. This paper propose a new hashing algorithm to overcome the storage space problem, while keeping a good accuracy and better query time. The Experimental results on a real large-scale dataset show the interest of our approach..

KW - Content based image retrieval (CBIR)

KW - Curse of dimensionality

KW - LSH

KW - Multidimensional indexing

KW - Nearest neighbour search

KW - Scalability

M3 - Papier

AN - SCOPUS:84983151072

SP - 303

EP - 318

Y2 - 18 March 2015 through 20 March 2015

ER -

SC-LSH: Une Méthode d'Indexation pour une Recherche de Similarité Approximative dans l'Espace Multidimensionnel

Abstract

Conference

Fingerprint

Cite this