InriaFBK at Germeval 2018: Identifying Offensive Tweets Using Recurrent Neural Networks

Michele Corazza; Stefano Menini; Pinar Arslan; Rachele Sprugnoli; Elena Cabrio; Sara Tonelli; Serena Villata

InriaFBK at Germeval 2018: Identifying Offensive Tweets Using Recurrent Neural Networks

Michele Corazza
, Stefano Menini
, Pinar Arslan
, Rachele Sprugnoli
, Elena Cabrio
, Sara Tonelli
, Serena Villata

Università Cattolica del Sacro Cuore

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

Abstract

In this paper, we describe two systems for predicting message-level offensive language in German tweets: one discriminates between offensive and not offensive messages, and the second performs a fine-grained classification by recognizing also classes of offense. Both systems are based on the same approach, which builds upon Recurrent Neural Networks used with the following features: word embeddings, emoji embeddings and social-network specific features. The model is able to combine word-level information and tweet-level information in order to perform the classification tasks.

Original language	English
Title of host publication	Proceedings of the GermEval 2018 Workshop
Pages	80-84
Number of pages	5
Publication status	Published - 2018
Event	GermEval 2018 - Vienna, Austria Duration: 21 Dec 2018 → 21 Dec 2018

Workshop

Workshop	GermEval 2018
City	Vienna, Austria
Period	21/12/18 → 21/12/18

Keywords

hate speech detection, neural networks, evaluation campaign

Cite this

@inproceedings{97e95cf7c16f4f7892a41a61dc314c1f,

title = "InriaFBK at Germeval 2018: Identifying Offensive Tweets Using Recurrent Neural Networks",

abstract = "In this paper, we describe two systems for predicting message-level offensive language in German tweets: one discriminates between offensive and not offensive messages, and the second performs a fine-grained classification by recognizing also classes of offense. Both systems are based on the same approach, which builds upon Recurrent Neural Networks used with the following features: word embeddings, emoji embeddings and social-network specific features. The model is able to combine word-level information and tweet-level information in order to perform the classification tasks.",

keywords = "hate speech detection, neural networks, evaluation campaign, hate speech detection, neural networks, evaluation campaign",

author = "Michele Corazza and Stefano Menini and Pinar Arslan and Rachele Sprugnoli and Elena Cabrio and Sara Tonelli and Serena Villata",

year = "2018",

language = "English",

pages = "80--84",

booktitle = "Proceedings of the GermEval 2018 Workshop",

note = "GermEval 2018 ; Conference date: 21-12-2018 Through 21-12-2018",

}

TY - GEN

T1 - InriaFBK at Germeval 2018: Identifying Offensive Tweets Using Recurrent Neural Networks

AU - Corazza, Michele

AU - Menini, Stefano

AU - Arslan, Pinar

AU - Sprugnoli, Rachele

AU - Cabrio, Elena

AU - Tonelli, Sara

AU - Villata, Serena

PY - 2018

Y1 - 2018

N2 - In this paper, we describe two systems for predicting message-level offensive language in German tweets: one discriminates between offensive and not offensive messages, and the second performs a fine-grained classification by recognizing also classes of offense. Both systems are based on the same approach, which builds upon Recurrent Neural Networks used with the following features: word embeddings, emoji embeddings and social-network specific features. The model is able to combine word-level information and tweet-level information in order to perform the classification tasks.

AB - In this paper, we describe two systems for predicting message-level offensive language in German tweets: one discriminates between offensive and not offensive messages, and the second performs a fine-grained classification by recognizing also classes of offense. Both systems are based on the same approach, which builds upon Recurrent Neural Networks used with the following features: word embeddings, emoji embeddings and social-network specific features. The model is able to combine word-level information and tweet-level information in order to perform the classification tasks.

KW - hate speech detection, neural networks, evaluation campaign

UR - http://hdl.handle.net/10807/133007

UR - https://www.oeaw.ac.at/fileadmin/subsites/academiaecorpora/pdf/germeval2018_proceedings.pdf

M3 - Conference contribution

SP - 80

EP - 84

BT - Proceedings of the GermEval 2018 Workshop

T2 - GermEval 2018

Y2 - 21 December 2018 through 21 December 2018

ER -

InriaFBK at Germeval 2018: Identifying Offensive Tweets Using Recurrent Neural Networks

Abstract

Workshop

Keywords

Other files and links

Fingerprint

Cite this