Using deep neural networks for extracting sentiment targets in arabic tweets

Ayman El-Kilany; Amr Azzam; Samhaa R. El-Beltagy

doi:10.1007/978-3-319-67056-0_1

Using deep neural networks for extracting sentiment targets in arabic tweets

Ayman El-Kilany, Amr Azzam, Samhaa R. El-Beltagy^*

^*Korrespondierende*r Autor*in für diese Arbeit

Publikation: Wissenschaftliche Fachzeitschrift › Originalbeitrag in Fachzeitschrift › Forschung › Begutachtung

Abstract

In this paper, we investigate the problem of recognizing entities which are targeted by text sentiment in Arabic tweets. To do so, we train a bidirectional LSTM deep neural network with conditional random fields as a classification layer on top of the network to discover the features of this specific set of entities and extract them from Arabic tweets. We’ve evaluated the network performance against a baseline method which makes use of a regular named entity recognizer and a sentiment analyzer. The deep neural network has shown a noticeable advantage in extracting sentiment target entities from Arabic tweets.

Originalsprache	Englisch
Seiten (von - bis)	3-15
Seitenumfang	13
Fachzeitschrift	Studies in Computational Intelligence
Jahrgang	740
DOIs	https://doi.org/10.1007/978-3-319-67056-0_1
Publikationsstatus	Veröffentlicht - 2018
Extern publiziert	Ja

Bibliographische Notiz

Funding Information:
Acknowledgements We would like to thank Abu Bakr Soliman for building the user interface used for annotating sentiment target data and for collecting the 54 million tweets that we have used for building our word embeddings model. This work was partially supported by ITIDA grant number [PRP]2015.R19.9.

Publisher Copyright:
© 2018, Springer International Publishing AG.

Zugriff auf Dokument

10.1007/978-3-319-67056-0_1

Verfügbarkeit prüfen

Katalog der Universitätsbibliothek

Andere Dateien und Links

Verknüpfung zur Publikation in Scopus

Zitat

@article{9734a07d5c2846cea5dab4d0b70dec5f,

title = "Using deep neural networks for extracting sentiment targets in arabic tweets",

abstract = "In this paper, we investigate the problem of recognizing entities which are targeted by text sentiment in Arabic tweets. To do so, we train a bidirectional LSTM deep neural network with conditional random fields as a classification layer on top of the network to discover the features of this specific set of entities and extract them from Arabic tweets. We{\textquoteright}ve evaluated the network performance against a baseline method which makes use of a regular named entity recognizer and a sentiment analyzer. The deep neural network has shown a noticeable advantage in extracting sentiment target entities from Arabic tweets.",

keywords = "Arabic, Conditional random fields, Long short-term memory networks, Recurrent neural networks, Sentiment target recognition",

author = "Ayman El-Kilany and Amr Azzam and El-Beltagy, {Samhaa R.}",

note = "Publisher Copyright: {\textcopyright} 2018, Springer International Publishing AG.",

year = "2018",

doi = "10.1007/978-3-319-67056-0_1",

language = "English",

volume = "740",

pages = "3--15",

journal = "Studies in Computational Intelligence",

issn = "1860-949X",

publisher = "Springer Verlag",

}

TY - JOUR

T1 - Using deep neural networks for extracting sentiment targets in arabic tweets

AU - El-Kilany, Ayman

AU - Azzam, Amr

AU - El-Beltagy, Samhaa R.

PY - 2018

Y1 - 2018

N2 - In this paper, we investigate the problem of recognizing entities which are targeted by text sentiment in Arabic tweets. To do so, we train a bidirectional LSTM deep neural network with conditional random fields as a classification layer on top of the network to discover the features of this specific set of entities and extract them from Arabic tweets. We’ve evaluated the network performance against a baseline method which makes use of a regular named entity recognizer and a sentiment analyzer. The deep neural network has shown a noticeable advantage in extracting sentiment target entities from Arabic tweets.

AB - In this paper, we investigate the problem of recognizing entities which are targeted by text sentiment in Arabic tweets. To do so, we train a bidirectional LSTM deep neural network with conditional random fields as a classification layer on top of the network to discover the features of this specific set of entities and extract them from Arabic tweets. We’ve evaluated the network performance against a baseline method which makes use of a regular named entity recognizer and a sentiment analyzer. The deep neural network has shown a noticeable advantage in extracting sentiment target entities from Arabic tweets.

KW - Arabic

KW - Conditional random fields

KW - Long short-term memory networks

KW - Recurrent neural networks

KW - Sentiment target recognition

UR - http://www.scopus.com/inward/record.url?scp=85034824988&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-67056-0_1

DO - 10.1007/978-3-319-67056-0_1

M3 - Journal article

AN - SCOPUS:85034824988

SN - 1860-949X

VL - 740

SP - 3

EP - 15

JO - Studies in Computational Intelligence

JF - Studies in Computational Intelligence

ER -