Know-Center at SemEval-2019 Task 5: Multilingual Hate SpeechDetection on Twitter using CNNs

Roman Kern; Kevin Winter

doi:10.18653/v1/S19-2076

Know-Center at SemEval-2019 Task 5: Multilingual Hate SpeechDetection on Twitter using CNNs

Roman Kern, Kevin Winter

Know-Center GmbH Research Center for Data-Driven Business & Big Data Analytics (98770)

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

Abstract

This paper presents the Know-Center system submitted for task 5 of the SemEval-2019 workshop. Given a Twitter message in either English or Spanish, the task is to first detect whether it contains hateful speech and second, to determine the target and level of aggression used. For this purpose our system utilizes word embeddings and a neural network architecture, consisting of both dilated and traditional convolution layers. We achieved average F1-scores of 0.57 and 0.74 for English and Spanish respectively.

Original language	English
Title of host publication	Proceedings of the Thirteenth International Workshop on Semantic Evaluation
Publisher	Association for Computational Linguistics
Pages	431-435
Number of pages	5
DOIs	https://doi.org/10.18653/v1/S19-2076
Publication status	Published - 2019
Event	13th International Workshop on Semantic Evaluation - Minneapolis, United States Duration: 6 Jun 2019 → 7 Jun 2019

Conference

Conference	13th International Workshop on Semantic Evaluation
Abbreviated title	SemEval 2019
Country/Territory	United States
City	Minneapolis
Period	6/06/19 → 7/06/19

Access to Document

10.18653/v1/S19-2076Licence: CC BY 4.0

https://www.aclweb.org/anthology/S19-2076.pdfLicence: CC BY 4.0

Cite this

Kern, R & Winter, K 2019, Know-Center at SemEval-2019 Task 5: Multilingual Hate SpeechDetection on Twitter using CNNs. in Proceedings of the Thirteenth International Workshop on Semantic Evaluation. Association for Computational Linguistics, pp. 431-435, 13th International Workshop on Semantic Evaluation, Minneapolis, Minnesota, United States, 6/06/19. https://doi.org/10.18653/v1/S19-2076

@inproceedings{e4fe5e960eaf407384c60cc565664323,

title = "Know-Center at SemEval-2019 Task 5: Multilingual Hate SpeechDetection on Twitter using CNNs",

abstract = "This paper presents the Know-Center system submitted for task 5 of the SemEval-2019 workshop. Given a Twitter message in either English or Spanish, the task is to first detect whether it contains hateful speech and second, to determine the target and level of aggression used. For this purpose our system utilizes word embeddings and a neural network architecture, consisting of both dilated and traditional convolution layers. We achieved average F1-scores of 0.57 and 0.74 for English and Spanish respectively.",

author = "Roman Kern and Kevin Winter",

year = "2019",

doi = "10.18653/v1/S19-2076",

language = "English",

pages = "431--435",

booktitle = "Proceedings of the Thirteenth International Workshop on Semantic Evaluation",

publisher = "Association for Computational Linguistics",

note = "13th International Workshop on Semantic Evaluation, SemEval 2019 ; Conference date: 06-06-2019 Through 07-06-2019",

}

TY - GEN

T1 - Know-Center at SemEval-2019 Task 5: Multilingual Hate SpeechDetection on Twitter using CNNs

AU - Kern, Roman

AU - Winter, Kevin

PY - 2019

Y1 - 2019

N2 - This paper presents the Know-Center system submitted for task 5 of the SemEval-2019 workshop. Given a Twitter message in either English or Spanish, the task is to first detect whether it contains hateful speech and second, to determine the target and level of aggression used. For this purpose our system utilizes word embeddings and a neural network architecture, consisting of both dilated and traditional convolution layers. We achieved average F1-scores of 0.57 and 0.74 for English and Spanish respectively.

AB - This paper presents the Know-Center system submitted for task 5 of the SemEval-2019 workshop. Given a Twitter message in either English or Spanish, the task is to first detect whether it contains hateful speech and second, to determine the target and level of aggression used. For this purpose our system utilizes word embeddings and a neural network architecture, consisting of both dilated and traditional convolution layers. We achieved average F1-scores of 0.57 and 0.74 for English and Spanish respectively.

U2 - 10.18653/v1/S19-2076

DO - 10.18653/v1/S19-2076

M3 - Conference paper

SP - 431

EP - 435

BT - Proceedings of the Thirteenth International Workshop on Semantic Evaluation

PB - Association for Computational Linguistics

T2 - 13th International Workshop on Semantic Evaluation

Y2 - 6 June 2019 through 7 June 2019

ER -