Overview of the GermEval 2018 Shared Task on the Identification of Offensive Language

By Michael Wiegand, Melanie Siegel. 2018

366 citations recorded by the source corpus.

Subfields: Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Language Generation Models Can Cause Harm: So What Can We Do About It? An Actionable Survey (2022, background): “Since language is inherently a tool of power—the primary means by *Equal contribution 1While the majority of these models are trained on English language datasets, recent studies have also obtained similar gains on other languages (Lin et al., 2021c; Shliazhko et al., 2022) which people and societie”
SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval) (2019, background): “Offensive language: The GermEval 2 (Wiegand et al., 2018) shared task focused on Offensive language identiﬁcation in German tweets.”
FLAIR: An Easy-to-Use Framework for State-of-the-Art NLP (2019, background): “, 2006) en default, fast NP Chunking CoNLL 2000 (Sang and Buchholz, 2000) en default, fast Offensive Language Detection GermEval 2018 (Wiegand et al., 2018) de default PoS tagging Ontonotes (Hovy et al.”
Predicting the Type and Target of Offensive Posts in Social Media (2019, background): “The GermEval 3 (Wiegand et al., 2018) shared task focused on Offensive language identiﬁcation in German tweets.”
Detection of Abusive Language: the Problem of Biased Datasets (2019, methodology): “On the set of abusive microposts of each dataset, we computed the proportion of microposts that include at least one abusive word according to the lexicon of abusive words from Wiegand et al. (2018a).”
SemEval-2019 Task 5: Multilingual Detection of Hate Speech Against Immigrants and Women in Twitter (2019, background): “Let us mention the ﬁrst and second edition of the Workshop on Abusive Language 1 (Waseem et al., 2017), the First Workshop on Trolling, Aggression and Cy-berbullying (Kumar et al., 2018), that also included a shared task on aggression identiﬁcation, the tracks on Automatic Misogyny Identiﬁcation (AM”
SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media (OffensEval 2020) (2020, background): “Given the multitude of terms and definitions related to offensive language used in the literature, several recent studies have investigated the common aspects of different abusive language detection tasks (Waseem et al., 2017; Wiegand et al., 2018).”
Resources and benchmark corpora for hate speech detection: a systematic review (2020, background): “Offensive language is the main track of OffensEval (Zampieri et al. 2019b, a) and the corresponding task at GermEval campaign in 2018 (Wiegand et al. 2018b).”
Toxicity Detection can be Sensitive to the Conversational Context (2021, background): “…of the published toxicity datasets contain posts in English, but datasets in other languages also exist, such as Greek (Pavlopoulos et al., 2017a), Arabic (Mubarak et al., 2017), French (Chiril et al., 2020), Indonesian (Ibrohim and Budi, 2018) and German (Ross et al., 2016; Wiegand et al., 2018).”
Towards multidomain and multilingual abusive language detection: a survey (2021, background): “Even if in the last years abusive language datasets were developed for other languages, including Italian [15, 41], Spanish [41], and German [148], English remains by far the most represented language.”
SOLID: A Large-Scale Semi-Supervised Dataset for Offensive Language Identification (2020, background): “There were several shared tasks that have focused speciﬁcally on offensive language identiﬁcation, e.g., GermEval-2018 (Wiegand et al., 2018), which focused on offensive language identiﬁcation in German tweets, HASOC-2019 (Mandl et al., 2019), and TRAC-2018(Fortuna et al., 2018).”
HateBERT: Retraining BERT for Abusive Language Detection in English (2020, methodology): “We selected all words with an offensiveness scores equal or higher than 0.75 from Wiegand et al. (2018)’s dictionary.”
German’s Next Language Model (2020, methodology): “For GermEval18 we report scores for the best-performing systems (Wiegand et al., 2018), and the result reported by FLAIR framework (Akbik et al.”
Hybrid Emoji-Based Masked Language Models for Zero-Shot Abusive Language Detection (2020, background): “GERMAN (Wiegand et al., 2018)”
Confronting Abusive Language Online: A Survey from the Ethical and Human Rights Perspective (2020, background): “For example, Wiegand et al. (2018) created a large lexicon and demonstrated its effectiveness in the cross-domain detection of abusive micro-posts.”
Tackling Online Abuse: A Survey of Automated Abuse Detection Methods (2019, methodology): “New abusive words and phrases continue to enter the language (Wiegand et al., 2018b).”