A Simple and Effective Scheme for Data Pre-processing in Extreme Classification

Sujay Khandagale; Rohit Babbar

A Simple and Effective Scheme for Data Pre-processing in Extreme Classification

Sujay Khandagale, Rohit Babbar

Tietotekniikan laitos

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference article in proceedings › Scientific › vertaisarvioitu

1 Sitaatiot (Scopus)

46 Lataukset (Pure)

Abstrakti

Extreme multi-label classification (XMC) refers to supervised multi-label learning involving hundreds of thousand or even millions of labels. It has been shown to be an effective framework for addressing crucial tasks such as recommendation, ranking and web-advertising. In this paper, we propose a method for effective and well-motivated data pre-processing scheme in XMC. We show that our proposed algorithm, PrunEX, can remove upto 90% data in the input which is redundant from a classification view-point. Our scheme is universal in the sense it is applicable to all known public datasets in the domain of XMC.

Alkuperäiskieli	Englanti
Otsikko	ESANN 2019 - Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Kustantaja	i6doc.com
Sivut	67-72
ISBN (painettu)	978-287-587-065-0
Tila	Julkaistu - 26 huhtik. 2019
OKM-julkaisutyyppi	A4 Artikkeli konferenssijulkaisussa
Tapahtuma	European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning - Bruges, Belgia Kesto: 24 huhtik. 2019 → 26 huhtik. 2019 Konferenssinumero: 27

Conference

Conference	European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Lyhennettä	ESANN
Maa/Alue	Belgia
Kaupunki	Bruges
Ajanjakso	24/04/2019 → 26/04/2019

Pääsy asiakirjaan

ESANNAccepted author manuscript, 276 KB

Muut tiedostot ja linkit

Science-IT
Hakala, M. (Manager)
Perustieteiden korkeakoulu
Laitteistot/tilat: Facility

Siteeraa tätä

@inproceedings{29699d7324744017b75bbeee61a9d274,

title = "A Simple and Effective Scheme for Data Pre-processing in Extreme Classification",

abstract = "Extreme multi-label classification (XMC) refers to supervised multi-label learning involving hundreds of thousand or even millions of labels. It has been shown to be an effective framework for addressing crucial tasks such as recommendation, ranking and web-advertising. In this paper, we propose a method for effective and well-motivated data pre-processing scheme in XMC. We show that our proposed algorithm, PrunEX, can remove upto 90\% data in the input which is redundant from a classification view-point. Our scheme is universal in the sense it is applicable to all known public datasets in the domain of XMC.",

author = "Sujay Khandagale and Rohit Babbar",

year = "2019",

month = apr,

day = "26",

language = "English",

isbn = "978-287-587-065-0",

pages = "67--72",

booktitle = "ESANN 2019 - Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

publisher = "i6doc.com",

address = "Belgium",

note = "European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN ; Conference date: 24-04-2019 Through 26-04-2019",

}

Khandagale, S & Babbar, R 2019, A Simple and Effective Scheme for Data Pre-processing in Extreme Classification. julkaisussa ESANN 2019 - Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com, Sivut 67-72, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, Bruges, Belgia, 24/04/2019.

A Simple and Effective Scheme for Data Pre-processing in Extreme Classification. / Khandagale, Sujay; Babbar, Rohit.
ESANN 2019 - Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com, 2019. s. 67-72.

Tutkimustuotos: Artikkeli kirjassa/konferenssijulkaisussa › Conference article in proceedings › Scientific › vertaisarvioitu

TY - GEN

T1 - A Simple and Effective Scheme for Data Pre-processing in Extreme Classification

AU - Khandagale, Sujay

AU - Babbar, Rohit

N1 - Conference code: 27

PY - 2019/4/26

Y1 - 2019/4/26

N2 - Extreme multi-label classification (XMC) refers to supervised multi-label learning involving hundreds of thousand or even millions of labels. It has been shown to be an effective framework for addressing crucial tasks such as recommendation, ranking and web-advertising. In this paper, we propose a method for effective and well-motivated data pre-processing scheme in XMC. We show that our proposed algorithm, PrunEX, can remove upto 90% data in the input which is redundant from a classification view-point. Our scheme is universal in the sense it is applicable to all known public datasets in the domain of XMC.

AB - Extreme multi-label classification (XMC) refers to supervised multi-label learning involving hundreds of thousand or even millions of labels. It has been shown to be an effective framework for addressing crucial tasks such as recommendation, ranking and web-advertising. In this paper, we propose a method for effective and well-motivated data pre-processing scheme in XMC. We show that our proposed algorithm, PrunEX, can remove upto 90% data in the input which is redundant from a classification view-point. Our scheme is universal in the sense it is applicable to all known public datasets in the domain of XMC.

UR - https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2019-167.pdf

UR - https://www.scopus.com/pages/publications/85071316674

M3 - Conference article in proceedings

SN - 978-287-587-065-0

SP - 67

EP - 72

BT - ESANN 2019 - Proceedings, 27th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

PB - i6doc.com

T2 - European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

Y2 - 24 April 2019 through 26 April 2019

ER -

A Simple and Effective Scheme for Data Pre-processing in Extreme Classification

Abstrakti

Conference

Pääsy asiakirjaan

Muut tiedostot ja linkit

Sormenjälki

Laitteet

Science-IT

Siteeraa tätä