Boosting Tricks for Word Mover’s Distance

Konstantinos Skianis; Fragkiskos D. Malliaros; Nikolaos Tziortziotis; Michalis Vazirgiannis

doi:10.1007/978-3-030-61616-8_61

Boosting Tricks for Word Mover’s Distance

Konstantinos Skianis, Fragkiskos D. Malliaros, Nikolaos Tziortziotis, Michalis Vazirgiannis

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Word embeddings have opened a new path in creating novel approaches for addressing traditional problems in the natural language processing (NLP) domain. However, using word embeddings to compare text documents remains a relatively unexplored topic—with Word Mover’s Distance (WMD) being the prominent tool used so far. In this paper, we present a variety of tools that can further improve the computation of distances between documents based on WMD. We demonstrate that, alternative stopwords, cross document-topic comparison, deep contextualized word vectors and convex metric learning, constitute powerful tools that can boost WMD.

Original language	English
Title of host publication	Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings
Editors	Igor Farkaš, Paolo Masulli, Stefan Wermter
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	761-772
Number of pages	12
ISBN (Print)	9783030616151
DOIs	https://doi.org/10.1007/978-3-030-61616-8_61
Publication status	Published - 1 Jan 2020
Event	29th International Conference on Artificial Neural Networks, ICANN 2020 - Bratislava, Slovakia Duration: 15 Sept 2020 → 18 Sept 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12397 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	29th International Conference on Artificial Neural Networks, ICANN 2020
Country/Territory	Slovakia
City	Bratislava
Period	15/09/20 → 18/09/20

Keywords

Text classification
Word embeddings
Word mover’s distance

Access to Document

10.1007/978-3-030-61616-8_61

Cite this

Skianis, K., Malliaros, F. D., Tziortziotis, N., & Vazirgiannis, M. (2020). Boosting Tricks for Word Mover’s Distance. In I. Farkaš, P. Masulli, & S. Wermter (Eds.), Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings (pp. 761-772). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12397 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-61616-8_61

Skianis, Konstantinos ; Malliaros, Fragkiskos D. ; Tziortziotis, Nikolaos et al. / Boosting Tricks for Word Mover’s Distance. Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings. editor / Igor Farkaš ; Paolo Masulli ; Stefan Wermter. Springer Science and Business Media Deutschland GmbH, 2020. pp. 761-772 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{95e3a73fa8c0467b92587c5967abd0ef,

title = "Boosting Tricks for Word Mover{\textquoteright}s Distance",

abstract = "Word embeddings have opened a new path in creating novel approaches for addressing traditional problems in the natural language processing (NLP) domain. However, using word embeddings to compare text documents remains a relatively unexplored topic—with Word Mover{\textquoteright}s Distance (WMD) being the prominent tool used so far. In this paper, we present a variety of tools that can further improve the computation of distances between documents based on WMD. We demonstrate that, alternative stopwords, cross document-topic comparison, deep contextualized word vectors and convex metric learning, constitute powerful tools that can boost WMD.",

keywords = "Text classification, Word embeddings, Word mover{\textquoteright}s distance",

author = "Konstantinos Skianis and Malliaros, \{Fragkiskos D.\} and Nikolaos Tziortziotis and Michalis Vazirgiannis",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 29th International Conference on Artificial Neural Networks, ICANN 2020 ; Conference date: 15-09-2020 Through 18-09-2020",

year = "2020",

month = jan,

day = "1",

doi = "10.1007/978-3-030-61616-8\_61",

language = "English",

isbn = "9783030616151",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "761--772",

editor = "Igor Farka{\v s} and Paolo Masulli and Stefan Wermter",

booktitle = "Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings",

}

Skianis, K, Malliaros, FD, Tziortziotis, N & Vazirgiannis, M 2020, Boosting Tricks for Word Mover’s Distance. in I Farkaš, P Masulli & S Wermter (eds), Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12397 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 761-772, 29th International Conference on Artificial Neural Networks, ICANN 2020, Bratislava, Slovakia, 15/09/20. https://doi.org/10.1007/978-3-030-61616-8_61

Boosting Tricks for Word Mover’s Distance. / Skianis, Konstantinos; Malliaros, Fragkiskos D.; Tziortziotis, Nikolaos et al.
Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings. ed. / Igor Farkaš; Paolo Masulli; Stefan Wermter. Springer Science and Business Media Deutschland GmbH, 2020. p. 761-772 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12397 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Boosting Tricks for Word Mover’s Distance

AU - Skianis, Konstantinos

AU - Malliaros, Fragkiskos D.

AU - Tziortziotis, Nikolaos

AU - Vazirgiannis, Michalis

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Word embeddings have opened a new path in creating novel approaches for addressing traditional problems in the natural language processing (NLP) domain. However, using word embeddings to compare text documents remains a relatively unexplored topic—with Word Mover’s Distance (WMD) being the prominent tool used so far. In this paper, we present a variety of tools that can further improve the computation of distances between documents based on WMD. We demonstrate that, alternative stopwords, cross document-topic comparison, deep contextualized word vectors and convex metric learning, constitute powerful tools that can boost WMD.

AB - Word embeddings have opened a new path in creating novel approaches for addressing traditional problems in the natural language processing (NLP) domain. However, using word embeddings to compare text documents remains a relatively unexplored topic—with Word Mover’s Distance (WMD) being the prominent tool used so far. In this paper, we present a variety of tools that can further improve the computation of distances between documents based on WMD. We demonstrate that, alternative stopwords, cross document-topic comparison, deep contextualized word vectors and convex metric learning, constitute powerful tools that can boost WMD.

KW - Text classification

KW - Word embeddings

KW - Word mover’s distance

U2 - 10.1007/978-3-030-61616-8_61

DO - 10.1007/978-3-030-61616-8_61

M3 - Conference contribution

AN - SCOPUS:85094160315

SN - 9783030616151

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 761

EP - 772

BT - Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings

A2 - Farkaš, Igor

A2 - Masulli, Paolo

A2 - Wermter, Stefan

PB - Springer Science and Business Media Deutschland GmbH

T2 - 29th International Conference on Artificial Neural Networks, ICANN 2020

Y2 - 15 September 2020 through 18 September 2020

ER -

Skianis K, Malliaros FD, Tziortziotis N, Vazirgiannis M. Boosting Tricks for Word Mover’s Distance. In Farkaš I, Masulli P, Wermter S, editors, Artificial Neural Networks and Machine Learning – ICANN 2020 - 29th International Conference on Artificial Neural Networks, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 761-772. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-61616-8_61

Boosting Tricks for Word Mover’s Distance

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this