Social media platforms allow users to freely share their opinions about issues or anything they feel like. However, they also make it easier to spread hate and abusive content. The Fulani ethnic group has been the victim of this unfortunate phenomenon. This paper introduces the HERDPhobia - the first annotated hate speech dataset on Fulani herders in Nigeria - in three languages: English, Nigerian-Pidgin, and Hausa. We present a benchmark experiment using pre-trained languages models to classify the tweets as either hateful or non-hateful. Our experiment shows that the XML-T model provides better performance with 99.83% weighted F1. We released the dataset at https://github.com/hausanlp/HERDPhobia for further research.

情感分析是NLP中研究最广泛的应用程序之一，但大多数工作都集中在具有大量数据的语言上。我们介绍了尼日利亚的四种口语最广泛的语言（Hausa，Igbo，Nigerian-Pidgin和Yor \'ub \'a）的第一个大规模的人类通知的Twitter情感数据集，该数据集由大约30,000个注释的推文组成（以及每种语言的大约30,000个）（以及14,000尼日利亚猎人），其中包括大量的代码混合推文。我们提出了文本收集，过滤，处理和标记方法，使我们能够为这些低资源语言创建数据集。我们评估了数据集上的预训练模型和转移策略。我们发现特定于语言的模型和语言适应性芬通常表现最好。我们将数据集，训练的模型，情感词典和代码释放到激励措施中，以代表性不足的语言进行情感分析。

Pre-training large neural language models, such as BERT, has led to impressive gains on many natural language processing (NLP) tasks. Although this method has proven to be effective for many domains, it might not always provide desirable benefits. In this paper, we study the effects of hateful pre-training on low-resource hate speech classification tasks. While previous studies on the English language have emphasized its importance, we aim to augment their observations with some non-obvious insights. We evaluate different variations of tweet-based BERT models pre-trained on hateful, non-hateful, and mixed subsets of a 40M tweet dataset. This evaluation is carried out for the Indian languages Hindi and Marathi. This paper is empirical evidence that hateful pre-training is not the best pre-training option for hate speech detection. We show that pre-training on non-hateful text from the target domain provides similar or better results. Further, we introduce HindTweetBERT and MahaTweetBERT, the first publicly available BERT models pre-trained on Hindi and Marathi tweets, respectively. We show that they provide state-of-the-art performance on hate speech classification tasks. We also release hateful BERT for the two languages and a gold hate speech evaluation benchmark HateEval-Hi and HateEval-Mr consisting of manually labeled 2000 tweets each. The models and data are available at https://github.com/l3cube-pune/MarathiNLP .

Hope is characterized as openness of spirit toward the future, a desire, expectation, and wish for something to happen or to be true that remarkably affects human's state of mind, emotions, behaviors, and decisions. Hope is usually associated with concepts of desired expectations and possibility/probability concerning the future. Despite its importance, hope has rarely been studied as a social media analysis task. This paper presents a hope speech dataset that classifies each tweet first into "Hope" and "Not Hope", then into three fine-grained hope categories: "Generalized Hope", "Realistic Hope", and "Unrealistic Hope" (along with "Not Hope"). English tweets in the first half of 2022 were collected to build this dataset. Furthermore, we describe our annotation process and guidelines in detail and discuss the challenges of classifying hope and the limitations of the existing hope speech detection corpora. In addition, we reported several baselines based on different learning approaches, such as traditional machine learning, deep learning, and transformers, to benchmark our dataset. We evaluated our baselines using weighted-averaged and macro-averaged F1-scores. Observations show that a strict process for annotator selection and detailed annotation guidelines enhanced the dataset's quality. This strict annotation process resulted in promising performance for simple machine learning classifiers with only bi-grams; however, binary and multiclass hope speech detection results reveal that contextual embedding models have higher performance in this dataset.

This paper presents work on detecting misogyny in the comments of a large Austrian German language newspaper forum. We describe the creation of a corpus of 6600 comments which were annotated with 5 levels of misogyny. The forum moderators were involved as experts in the creation of the annotation guidelines and the annotation of the comments. We also describe the results of training transformer-based classification models for both binarized and original label classification of that corpus.

仇恨言论等攻击性内容的广泛构成了越来越多的社会问题。 AI工具是支持在线平台的审核过程所必需的。为了评估这些识别工具，需要与不同语言的数据集进行连续实验。 HASOC轨道（仇恨语音和冒犯性内容识别）专用于为此目的开发基准数据。本文介绍了英语，印地语和马拉地赛的Hasoc Subtrack。数据集由Twitter组装。此子系统有两个子任务。任务A是为所有三种语言提供的二进制分类问题（仇恨而非冒犯）。任务B是三个课程（仇恨）仇恨言论，令人攻击和亵渎为英语和印地语提供的细粒度分类问题。总体而言，652名队伍提交了652次。任务A最佳分类算法的性能分别为Marathi，印地语和英语的0.91,0.78和0.83尺寸。此概述介绍了任务和数据开发以及详细结果。提交竞争的系统应用了各种技术。最好的表演算法主要是变压器架构的变种。

语言随着时间的流逝而演变，单词含义会发生相应的变化。在社交媒体中尤其如此，因为它的动态性质会导致语义转移的速度更快，这使得NLP模型在处理新内容和趋势方面具有挑战性。但是，专门解决这些社交平台动态性质的数据集和模型的数量很少。为了弥合这一差距，我们提出了Tempowic，这是一种新的基准，尤其是旨在加快基于社交媒体的含义转变的研究。我们的结果表明，即使对于最近发行的专门从事社交媒体的语言模型，Tempowic是一个具有挑战性的基准。

Following the outbreak of a global pandemic, online content is filled with hate speech. Donald Trump's ''Chinese Virus'' tweet shifted the blame for the spread of the Covid-19 virus to China and the Chinese people, which triggered a new round of anti-China hate both online and offline. This research intends to examine China-related hate speech on Twitter during the two years following the burst of the pandemic (2020 and 2021). Through Twitter's API, in total 2,172,333 tweets hashtagged #china posted during the time were collected. By employing multiple state-of-the-art pretrained language models for hate speech detection, we identify a wide range of hate of various types, resulting in an automatically labeled anti-China hate speech dataset. We identify a hateful rate in #china tweets of 2.5% in 2020 and 1.9% in 2021. This is well above the average rate of online hate speech on Twitter at 0.6% identified in Gao et al., 2017. We further analyzed the longitudinal development of #china tweets and those identified as hateful in 2020 and 2021 through visualizing the daily number and hate rate over the two years. Our keyword analysis of hate speech in #china tweets reveals the most frequently mentioned terms in the hateful #china tweets, which can be used for further social science studies.

我们介绍了仇恨言论推文的Hateval语料库（Basile等，2019年）的丰富，旨在促进自动化的反叙事一代。与以前的工作相比（Chung etal。2019），手动书面反叙事与推文有关。但是，仅此信息似乎不足以获得反叙事生成的令人满意的语言模型。这就是为什么我们还根据Wagemanns（2016）提供了带有争论性信息的注释推文，我们认为可以帮助建立令人信服和有效的反叙事，以针对特定群体进行仇恨言论。我们讨论了这种注释过程的充分和困难，并提出了几个基线以自动检测带注释的元素。初步结果表明，自动注释者会靠近人类注释者来检测论证的某些方面，而其他人仅达到低或中等水平的通知者一致性。

在线仇恨言语检测已随着数字设备的增长而变得重要，但是英语以外的其他语言资源非常有限。我们介绍了K-MHAS，这是一种新的多标签数据集，用于仇恨言语检测，可有效处理韩国语言模式。该数据集由新闻评论中的109k话语组成，并提供了从1到4个标签的多标签分类，并处理主观性和相交性。我们评估了K-MHAS上强的基线。Kr-Bert带有子字符的代币器优于表现，在每个仇恨言论类中都认识到分解的角色。

仇恨言语检测模型通常在持有的测试集上评估。但是，这有可能因为仇恨言语数据集中越来越有据可查的系统差距和偏见，因此绘制模型性能的不完整且潜在的误导性图片。为了实现更多针对性的诊断见解，最近的研究引入了仇恨言语检测模型的功能测试。但是，这些测试目前仅针对英语内容，这意味着它们无法支持全球数十亿语言所说的其他语言中更有效模型的开发。为了帮助解决这个问题，我们介绍了多语言Hatecheck（MHC），这是一套用于多语言仇恨言语检测模型的功能测试。 MHC涵盖了跨十种语言的34个功能，这比任何其他仇恨语音数据集更多。为了说明MHC的效用，我们训练和测试了高性能的多语言仇恨语音检测模型，并揭示了单语和跨语性应用的关键模型弱点。

Due to the severity of the social media offensive and hateful comments in Brazil, and the lack of research in Portuguese, this paper provides the first large-scale expert annotated corpus of Brazilian Instagram comments for hate speech and offensive language detection. The HateBR corpus was collected from the comment section of Brazilian politicians' accounts on Instagram and manually annotated by specialists, reaching a high inter-annotator agreement. The corpus consists of 7,000 documents annotated according to three different layers: a binary classification (offensive versus non-offensive comments), offensiveness-level classification (highly, moderately, and slightly offensive), and nine hate speech groups (xenophobia, racism, homophobia, sexism, religious intolerance, partyism, apology for the dictatorship, antisemitism, and fatphobia). We also implemented baseline experiments for offensive language and hate speech detection and compared them with a literature baseline. Results show that the baseline experiments on our corpus outperform the current state-of-the-art for the Portuguese language.

仇恨言语检测模型的性能取决于对模型的训练数据集。现有的数据集大部分是由有限数量的实例或定义仇恨主题的仇恨域准备的。这阻碍了关于仇恨领域的大规模分析和转移学习。在这项研究中，我们构建了大规模的推文数据集，以用英语和低资源语言（土耳其语）进行仇恨言论检测，每个人都由每个标签的100k推文组成。我们的数据集设计为在五个域上分布的推文数量相等。统计测试支持的实验结果表明，基于变压器的语言模型的表现优于传统词袋和神经模型的英语至少5％，而土耳其语则优于大规模仇恨言语检测。该性能也可扩展到不同的训练规模，在使用20％的培训实例时，将回收98％的英语表现和土耳其语的97％。我们进一步研究了仇恨领域之间跨域转移的概括能力。我们表明，其他英语域平均有96％的目标域性能恢复，而土耳其语为92％。性别和宗教更成功地概括到其他领域，而体育运动最大。

Covid-19-Pandemic继续在社交媒体上提出各种讨论或辩论的主题。为了探索大流行对人们生活的影响，了解公众对与大流行有关的实体（例如药物，疫苗）对社交媒体的关注和态度至关重要。但是，对现有命名实体识别（NER）或目标情感分析（TSA）数据集培训的模型具有有限的理解与COVID相关的社交媒体文本的能力有限，因为这些数据集并未从医学角度设计或注释。本文释放了Mets-COV，这是一种包含医疗实体的数据集和与COVID相关的推文中的目标情感。 Mets-COV包含10,000条带有7种实体的推文，包括4种医疗实体类型（疾病，药物，症状和疫苗）和3种通用实体类型（人，位置和组织）。为了进一步调查推文用户对特定实体的态度，选择了4种类型的实体（人，组织，药物和疫苗），并用用户情感注释，从而产生了具有9,101个实体（5,278个推文）的目标情感数据集。据我们所知，METS-COV是第一个收集与COVID相关推文的医疗实体和相应情感的数据集。我们通过广泛的实验对经典机器学习模型和最先进的深度学习模型进行基准测试。结果表明，该数据集在NER和TSA任务方面都有大量改进的空间。 METS-COV是开发更好的医学社交媒体工具并促进计算社会科学研究的重要资源，尤其是在流行病学方面。我们的数据，注释准则，基准模型和源代码公开可用（https://github.com/ylab-open/mets-cov），以确保可重复性。

Recent directions for offensive language detection are hierarchical modeling, identifying the type and the target of offensive language, and interpretability with offensive span annotation and prediction. These improvements are focused on English and do not transfer well to other languages because of cultural and linguistic differences. In this paper, we present the Korean Offensive Language Dataset (KOLD) comprising 40,429 comments, which are annotated hierarchically with the type and the target of offensive language, accompanied by annotations of the corresponding text spans. We collect the comments from NAVER news and YouTube platform and provide the titles of the articles and videos as the context information for the annotation process. We use these annotated comments as training data for Korean BERT and RoBERTa models and find that they are effective at offensiveness detection, target classification, and target span detection while having room for improvement for target group classification and offensive span detection. We discover that the target group distribution differs drastically from the existing English datasets, and observe that providing the context information improves the model performance in offensiveness detection (+0.3), target classification (+1.5), and target group classification (+13.1). We publicly release the dataset and baseline models.

Automated offensive language detection is essential in combating the spread of hate speech, particularly in social media. This paper describes our work on Offensive Language Identification in low resource Indic language Marathi. The problem is formulated as a text classification task to identify a tweet as offensive or non-offensive. We evaluate different mono-lingual and multi-lingual BERT models on this classification task, focusing on BERT models pre-trained with social media datasets. We compare the performance of MuRIL, MahaTweetBERT, MahaTweetBERT-Hateful, and MahaBERT on the HASOC 2022 test set. We also explore external data augmentation from other existing Marathi hate speech corpus HASOC 2021 and L3Cube-MahaHate. The MahaTweetBERT, a BERT model, pre-trained on Marathi tweets when fine-tuned on the combined dataset (HASOC 2021 + HASOC 2022 + MahaHate), outperforms all models with an F1 score of 98.43 on the HASOC 2022 test set. With this, we also provide a new state-of-the-art result on HASOC 2022 / MOLD v2 test set.

在线仇恨是许多社交媒体平台的日益关注。为解决此问题，不同的社交媒体平台为此类内容引入了审核策略。他们还聘请了可以检查职位违反审议政策的职位并采取适当行动。辱骂语言研究领域的院士也进行各种研究以更好地检测此类内容。虽然在英语中有广泛的辱骂语言检测，但在这场火灾中，在印度，乌尔都语等低资源语言中有一个滥用语言检测的空格。在URDU中提出滥用语言检测数据集以及威胁性语言检测。在本文中，我们探索了XGBoost，LGBM，基于M-BERT的M-BERT模型的多种机器学习模型，用于基于共享任务的URDU滥用和威胁的内容检测。我们观察了在阿拉伯语中滥用语言数据集的变压器模型有助于获得最佳性能。我们的模型首先是滥用和威胁性的内容检测，分别使用0.88和0.54的F1Scoreof。

猖獗在社交媒体上使用令人反感的语言导致最近努力自动识别这种语言。虽然令人反感的语言具有一般特征，但对特定实体的攻击可能表现出明显的现象，例如名称拼写中的恶意改变。在本文中，我们提出了一种识别实体特定攻击性语言的方法。我们雇用了两个关键洞察力，即Twitter上的回复通常意味着反对派，一些账户持续到他们对特定目标的攻击性。使用我们的方法，我们能够收集数千名有针对性的令人反感的推文。我们在使用基于深度学习和支持矢量机基础的分类器时，我们展示了在阿拉伯语推文上具有13％和79％的相对F1措施改进的阿拉伯语推文的效果。此外，通过在多个实体上自动识别的攻击性推文扩展训练集可以提高F1-Measure48％。

自2020年初以来，Covid-19-19造成了全球重大影响。这给社会带来了很多困惑，尤其是由于错误信息通过社交媒体传播。尽管已经有几项与在社交媒体数据中发现错误信息有关的研究，但大多数研究都集中在英语数据集上。印度尼西亚的COVID-19错误信息检测的研究仍然很少。因此，通过这项研究，我们收集和注释印尼语的数据集，并通过考虑该推文的相关性来构建用于检测COVID-19错误信息的预测模型。数据集构造是由一组注释者进行的，他们标记了推文数据的相关性和错误信息。在这项研究中，我们使用印度培训预培训的语言模型提出了两阶段分类器模型，以进行推文错误信息检测任务。我们还尝试了其他几种基线模型进行文本分类。实验结果表明，对于相关性预测，BERT序列分类器的组合和用于错误信息检测的BI-LSTM的组合优于其他机器学习模型，精度为87.02％。总体而言，BERT利用率有助于大多数预测模型的更高性能。我们发布了高质量的Covid-19错误信息推文语料库，用高通道一致性表示。

We investigate how annotators' insensitivity to differences in dialect can lead to racial bias in automatic hate speech detection models, potentially amplifying harm against minority populations. We first uncover unexpected correlations between surface markers of African American English (AAE) and ratings of toxicity in several widely-used hate speech datasets. Then, we show that models trained on these corpora acquire and propagate these biases, such that AAE tweets and tweets by self-identified African Americans are up to two times more likely to be labelled as offensive compared to others. Finally, we propose dialect and race priming as ways to reduce the racial bias in annotation, showing that when annotators are made explicitly aware of an AAE tweet's dialect they are significantly less likely to label the tweet as offensive.