GSI-UPM at SemEval-2019 Task 5: Semantic Similarity and Word Embeddings for Multilingual Detection of Hate Speech Against Immigrants and Women on Twitter

Proceedings of the 13th International Workshop on Semantic Evaluation (SemEval-2019), pages 396–403 Minneapolis, Minnesota, USA, June 6–7, 2019. ©2019 Association for Computational Linguistics 396 GSI-UPM at SemEval-2019 Task 5: Semantic Similarity and Word Embeddings for Multilingual Detection of Hate Speech Against Immigrants and Women on Twitter Diego Benito, Oscar Araque, and Carlos A. Iglesias Intelligent Systems Group Universidad Polit´ecnica de Madrid Madrid, Spain Avenida Complutense, 30 {d.benito,o.araque,carlosangel.iglesias}@upm.es Abstract This paper describes the GSI-UPM system for SemEval-2019 Task 5, which tackles multi- lingual detection of hate speech on Twitter. The main contribution of the paper is the use of a method based on word embeddings and semantic similarity combined with traditional paradigms, such as n-grams, TF-IDF and POS. This combination of several features is fine- tuned through ablation tests, demonstrating the usefulness of different features. While our ap- proach outperforms baseline classifiers on dif- ferent sub-tasks, the best of our submitted runs reached the 5th position on the Spanish sub- task A. 1 Introduction Information available in social networks is the re- sult of many interactions between users and their activity on the net. Unfortunately, hate speech and other misuses are proliferating on the Internet. Hate speech authors justify their conduct based on the freedom of speech argument. Thus, a debate over hate speech legislation and freedom of speech has been generated (Herz and Molnar, 2012). The task to decide if a piece of text contains hate speech is not trivial, even for humans. Being subject to different interpretations and opinions, the manifestations of hate speech become difficult to define. Based on previous hate speech state- ments (Fortuna and Nunes, 2018; Schmidt and Wiegand, 2017), this phenomenon could be de- fined as offensive or humorist content in form of text, video, or images that attacks, diminishes, in- cites violence or hate against groups or individu- als, based on actual or perceived specific charac- teristics such as physical appearance, religion, de- scent, national or ethnic origin, sexual orientation, gender identity, or any other. Hate speech topic has gained impact and popu- larity in recent years, which is reflected not only by the increased media coverage but also by the growing political attention. Regarding the specific forms of hate speech that we deal with, sexism and racism victims increased during 2017 according to the FBI hate crime statistics1. For this reason, participating in SemEval2019 Task 5 (Basile et al., 2019) is such an interest- ing challenge. The proposed task consists in Hate Speech detection in Twitter messages featured by two specific different targets intrinsically related to the phenomena mentioned above, immigrants and women. The task is enriched by adding a multilingual perspective fostering the research for both English and Spanish messages. The system proposed relies on a supervised classifier using different text features combined with several strategies with the aim of finding an optimal performance. The remainder of this paper is structured as follows. After this introductory section, Section 2 reviews related work. Follow- ing, the proposed classification model is described in Section …

Content truncated to 500 words