ltl.uni-due at SemEval-2019 Task 5: Simple but Effective Lexico-Semantic Features for Detecting Hate Speech in Twitter

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 441–446 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 441 ltl.uni-due at SemEval-2019 Task 5: Simple but Effective Lexico-Semantic Features for Detecting Hate Speech in Twitter Huangpan Zhang, Michael Wojatzki, Tobias Horsmann and Torsten Zesch Language Technology Lab, University of Duisburg-Essen {huangpan.zhang, michael.wojatzki}@uni-due.de {tobias.horsmann, torsten.zesch}@uni-due.de Abstract In this paper, we present our contribution to SemEval 2019 Task 5 Multilingual Detection of Hate, specifically in the Subtask A (En- glish and Spanish). We compare different configurations of shallow and deep learning approaches on the English data and use the system that performs best in both sub-tasks. The resulting SVM-based system with lexico- semantic features (n-grams and embeddings) is ranked 23rd out of 69 on the English data and beats the baseline system. On the Spanish data our system is ranked 25th out of 39. 1 Introduction Hateful, abusive, or offending statements which target individuals or groups on the basis of char- acteristics such as gender, nationality, or sexual orientation are called hate speech (Basile et al., 2019). Social media is particularly affected by hate speech, as it is known to poison the com- munication climate, build up negative sentiment towards groups of people, or even lead to real- life consequences (Warner and Hirschberg, 2012; Waseem and Hovy, 2016; Schmidt and Wiegand, 2017; Wojatzki et al., 2018; Benikova et al., 2017; Ross et al., 2017). In this work, we present our submission to the SemEval 2019 Task 5: Multilingual Detection of Hate (Subtask A) for English and Spanish. The objective in Subtask A was to build a system which is able to predict whether given tweets in English or in Spanish are hateful or not hateful to- wards women or immigrants. We develop a hate speech detection system by experimenting with a range of classifiers which are either based on engineered features or on neu- ral network architectures. We systematically com- pare the performance of these different detection systems and for our final submission (for both En- glish and Spanish) we use the model that performs best on the English training data. Our best sys- tem is a SVM equipped with n-gram features and fastText (Mikolov et al., 2018) embeddings. Our system obtains the 23rd rank (out of 69) on the En- glish dataset and the 25th rank (out of 39) on the Spanish dataset. 2 System Description For our submission, we compare a wide range of different neural and non-neural systems in terms of their performance. Our actual submission sys- tem is the system that performed best in this eval- uation. We will now describe both our neural net- work approaches and the feature-engineering ap- proaches for detecting whether tweets are hateful towards women or immigrants (Subtask A). We developed and evaluated the approaches for the English dataset and applied the best performing system as-is to the Spanish data. We now first briefly describe the provided data and then discuss the prediction approaches in more detail. Dataset …

Content truncated to 500 words