HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks

Lidia Pivovarova; Llorenc Escoter; Arto Klami; Roman Yangarber

doi:10.18653/v1/s17-2143

HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks

Lidia Pivovarova, Llorenc Escoter, Arto Klami, Roman Yangarber

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

Sammanfattning

Task 5 of SemEval-2017 involves fine-grained sentiment analysis on financial
microblogs and news. Our solution for determining the sentiment score extends
an earlier convolutional neural network for sentiment analysis in several ways.
We explicitly encode a focus on a particular company, we apply a data
augmentation scheme, and use a larger data collection to complement the small
training data provided by the task organizers. The best results were achieved
by training a model on an external dataset and then tuning it using the
provided training dataset.

Originalspråk	engelska
Titel på värdpublikation	11th International Workshop on Semantic Evaluations (SemEval-2017) : Proceedings of the Workshop
Antal sidor	5
Utgivningsort	Stroudsburg, PA
Förlag	The Association for Computational Linguistics
Utgivningsdatum	aug. 2017
Sidor	842-846
ISBN (elektroniskt)	978-1-945626-55-5
DOI	https://doi.org/10.18653/v1/s17-2143
Status	Publicerad - aug. 2017
MoE-publikationstyp	A4 Artikel i en konferenspublikation
Evenemang	International Workshop on Semantic Evaluations - New Orleans, Förenta Staterna (USA) Varaktighet: 5 juni 2018 → 6 juni 2018 Konferensnummer: 12 http://alt.qcri.org/semeval2018/

Vetenskapsgrenar

113 Data- och informationsvetenskap

Åtkomst av dokument

10.18653/v1/s17-2143

HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks

LLL: Language Learning Lab
Yangarber, R. (Projektledare), Katinskaia, A. (Deltagare), Hou, J. (Deltagare), Furlan, G. (Deltagare) & Kylliäinen, I. P. (Deltagare)
Projekt: Forskningsprojekt

Citera det här

@inproceedings{cc4bafa5513e47b9b82e268272a55d87,

title = "HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks",

abstract = "Task 5 of SemEval-2017 involves fine-grained sentiment analysis on financialmicroblogs and news. Our solution for determining the sentiment score extendsan earlier convolutional neural network for sentiment analysis in several ways. We explicitly encode a focus on a particular company, we apply a dataaugmentation scheme, and use a larger data collection to complement the smalltraining data provided by the task organizers. The best results were achieved by training a model on an external dataset and then tuning it using the provided training dataset.",

keywords = "113 Computer and information sciences",

author = "Lidia Pivovarova and Llorenc Escoter and Arto Klami and Roman Yangarber",

year = "2017",

month = aug,

doi = "10.18653/v1/s17-2143",

language = "English",

pages = "842--846",

booktitle = "11th International Workshop on Semantic Evaluations (SemEval-2017)",

publisher = "The Association for Computational Linguistics",

address = "United States",

note = "International Workshop on Semantic Evaluations ; Conference date: 05-06-2018 Through 06-06-2018",

url = "http://alt.qcri.org/semeval2018/",

}

Pivovarova, L, Escoter, L, Klami, A & Yangarber, R 2017, HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks. i 11th International Workshop on Semantic Evaluations (SemEval-2017): Proceedings of the Workshop. The Association for Computational Linguistics, Stroudsburg, PA, s. 842-846, International Workshop on Semantic Evaluations, New Orleans, Förenta Staterna (USA), 05/06/2018. https://doi.org/10.18653/v1/s17-2143

HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks. / Pivovarova, Lidia; Escoter, Llorenc; Klami, Arto et al.
11th International Workshop on Semantic Evaluations (SemEval-2017): Proceedings of the Workshop. Stroudsburg, PA: The Association for Computational Linguistics, 2017. s. 842-846.

Forskningsoutput: Kapitel i bok/rapport/konferenshandling › Konferensbidrag › Vetenskaplig › Peer review

TY - GEN

T1 - HCS at SemEval-2017 Task 5

T2 - International Workshop on Semantic Evaluations

AU - Pivovarova, Lidia

AU - Escoter, Llorenc

AU - Klami, Arto

AU - Yangarber, Roman

N1 - Conference code: 12

PY - 2017/8

Y1 - 2017/8

N2 - Task 5 of SemEval-2017 involves fine-grained sentiment analysis on financialmicroblogs and news. Our solution for determining the sentiment score extendsan earlier convolutional neural network for sentiment analysis in several ways. We explicitly encode a focus on a particular company, we apply a dataaugmentation scheme, and use a larger data collection to complement the smalltraining data provided by the task organizers. The best results were achieved by training a model on an external dataset and then tuning it using the provided training dataset.

AB - Task 5 of SemEval-2017 involves fine-grained sentiment analysis on financialmicroblogs and news. Our solution for determining the sentiment score extendsan earlier convolutional neural network for sentiment analysis in several ways. We explicitly encode a focus on a particular company, we apply a dataaugmentation scheme, and use a larger data collection to complement the smalltraining data provided by the task organizers. The best results were achieved by training a model on an external dataset and then tuning it using the provided training dataset.

KW - 113 Computer and information sciences

U2 - 10.18653/v1/s17-2143

DO - 10.18653/v1/s17-2143

M3 - Conference contribution

SP - 842

EP - 846

BT - 11th International Workshop on Semantic Evaluations (SemEval-2017)

PB - The Association for Computational Linguistics

CY - Stroudsburg, PA

Y2 - 5 June 2018 through 6 June 2018

ER -

HCS at SemEval-2017 Task 5: Sentiment Detection in Business News Using Convolutional Neural Networks

Sammanfattning

Vetenskapsgrenar

Åtkomst av dokument

Projekt

LLL: Language Learning Lab

Citera det här