Memory-based Named Entity Recognition in Tweets

Antal Van den Bosch; Toine Bogers

Memory-based Named Entity Recognition in Tweets

Antal Van den Bosch, Toine Bogers

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

Abstract

We present a memory-based named entity recognition system that participated in the MSM-2013 Concept Extraction Challenge. The system expands the training set of annotated tweets with part-of-speech tags and seedlist information, and then generates a sequential memory-based tagger comprised of separate modules for known and unknown words. Two taggers are trained: one on the original capitalized data, and one on a lowercased version of the training data. The intersection of named entities in the predictions of the two taggers is kept as the final output.

Originalsprog	Engelsk
Titel	MSM 2013 : Proceedings of the 3rd WWW Workshop on Making Sense of Microposts
Redaktører	Amparo Cano, Matthew Rowe, Milan Stankovic, Aba-Sah Dadzie
Antal sider	4
Forlag	CEUR Workshop Proceedings
Publikationsdato	13 maj 2013
Sider	40-43
Status	Udgivet - 13 maj 2013
Udgivet eksternt	Ja
Begivenhed	3rd workshop on 'Making Sense of Microposts' - RIo de Janerio, Brasilien Varighed: 13 maj 2013 → … http://oak.dcs.shef.ac.uk/msm2013/

Konference

Konference	3rd workshop on 'Making Sense of Microposts'
Land/Område	Brasilien
By	RIo de Janerio
Periode	13/05/2013 → …
Internetadresse	http://oak.dcs.shef.ac.uk/msm2013/

Navn	CEUR Workshop Proceedings
Vol/bind	1019
ISSN	1613-0073

Adgang til dokumentet

msm_workshop_2013Forlagets udgivne version, 314 KB

Citationsformater

@inproceedings{a546dac0dbeb4e39bb97e8efaf01fd17,

title = "Memory-based Named Entity Recognition in Tweets",

abstract = "We present a memory-based named entity recognition system that participated in the MSM-2013 Concept Extraction Challenge. The system expands the training set of annotated tweets with part-of-speech tags and seedlist information, and then generates a sequential memory-based tagger comprised of separate modules for known and unknown words. Two taggers are trained: one on the original capitalized data, and one on a lowercased version of the training data. The intersection of named entities in the predictions of the two taggers is kept as the final output.",

keywords = "memory-based named entity recognition, concept extraction, natural language processing, sequential tagger, annotated tweets",

author = "{Van den Bosch}, Antal and Toine Bogers",

year = "2013",

month = may,

day = "13",

language = "English",

series = "CEUR Workshop Proceedings",

publisher = "CEUR Workshop Proceedings",

pages = "40--43",

editor = "Amparo Cano and Matthew Rowe and Milan Stankovic and Aba-Sah Dadzie",

booktitle = "MSM 2013",

note = "3rd workshop on 'Making Sense of Microposts', MSM2013 ; Conference date: 13-05-2013",

url = "http://oak.dcs.shef.ac.uk/msm2013/",

}

Van den Bosch, A & Bogers, T 2013, Memory-based Named Entity Recognition in Tweets. i A Cano, M Rowe, M Stankovic & A-S Dadzie (red), MSM 2013: Proceedings of the 3rd WWW Workshop on Making Sense of Microposts. CEUR Workshop Proceedings, CEUR Workshop Proceedings, bind 1019, s. 40-43, 3rd workshop on 'Making Sense of Microposts', RIo de Janerio, Brasilien, 13/05/2013. <http://oak.dcs.shef.ac.uk/msm2013/>

Memory-based Named Entity Recognition in Tweets. / Van den Bosch, Antal; Bogers, Toine.
MSM 2013: Proceedings of the 3rd WWW Workshop on Making Sense of Microposts. red. / Amparo Cano; Matthew Rowe; Milan Stankovic; Aba-Sah Dadzie. CEUR Workshop Proceedings, 2013. s. 40-43 (CEUR Workshop Proceedings, Bind 1019).

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Memory-based Named Entity Recognition in Tweets

AU - Van den Bosch, Antal

AU - Bogers, Toine

PY - 2013/5/13

Y1 - 2013/5/13

N2 - We present a memory-based named entity recognition system that participated in the MSM-2013 Concept Extraction Challenge. The system expands the training set of annotated tweets with part-of-speech tags and seedlist information, and then generates a sequential memory-based tagger comprised of separate modules for known and unknown words. Two taggers are trained: one on the original capitalized data, and one on a lowercased version of the training data. The intersection of named entities in the predictions of the two taggers is kept as the final output.

AB - We present a memory-based named entity recognition system that participated in the MSM-2013 Concept Extraction Challenge. The system expands the training set of annotated tweets with part-of-speech tags and seedlist information, and then generates a sequential memory-based tagger comprised of separate modules for known and unknown words. Two taggers are trained: one on the original capitalized data, and one on a lowercased version of the training data. The intersection of named entities in the predictions of the two taggers is kept as the final output.

KW - memory-based named entity recognition

KW - concept extraction

KW - natural language processing

KW - sequential tagger

KW - annotated tweets

M3 - Article in proceedings

T3 - CEUR Workshop Proceedings

SP - 40

EP - 43

BT - MSM 2013

A2 - Cano, Amparo

A2 - Rowe, Matthew

A2 - Stankovic, Milan

A2 - Dadzie, Aba-Sah

PB - CEUR Workshop Proceedings

T2 - 3rd workshop on 'Making Sense of Microposts'

Y2 - 13 May 2013

ER -

Memory-based Named Entity Recognition in Tweets

Abstract

Konference

Adgang til dokumentet

Fingeraftryk

Citationsformater