Search

Hate Speech already been phenomenal expansion over the past decade. The paper proposed a new model that combines advanced CNN and Bidirectional Encoder Representations from Transformers (BERT) context embedding to predict hate speech in social media. This research trained contextual embedding on the datasets and used the learned information to identify objectionable language and hate speech in text. The paper evaluated supervised machine learning classifiers for bigoted and offensive content on Twitter using two datasets and found that advanced CNN context embeddings produced superior results. This research generated optimistic outcomes, which achieves 73% F1-score for Davidson dataset and 56% F1-score for TRAC-1 dataset [ABSTRACT FROM AUTHOR]

Hate speech is a public statement that demonstrates resentment or provokes disturbance towards a person or group often based upon race, age, religion, sexual orientation, minority group, psychical disability, political belief, etc. Such an act is a leading cause of mental degeneration in individuals observed throughout the world. We have witnessed an upsurge in the spreading of hateful speech through videos in recent times due to increased social media usage. Researchers are working on this issue because it has become more frequent on several social media platforms, and it leads to low self-esteem and has significant negative impacts on human life. In this work, we focus on collecting data from such videos as nowadays people are sharing numerous videos of this negative nature on platforms like Facebook and YouTube. The audio data of these videos were then converted into text to build the dataset, and we applied some classifier models to our dataset. In this paper, we utilized a transfer learning Bidirectional Encoder Representations from Transformers (BERT) model that gives state-of-the-art outcomes. More precisely, we fine-tuned our model based on transfer learning to evaluate BERT's capacity to capture hostile contexts inside YouTube videos. We examined Fine–Tuning BERT; with different learning rates and listed the outcomes. We train the BERT by freezing all the hyperparameters but with various random seeds to evaluate our suggested Fine-tuning approach. Compared to previous methodologies that used our dataset, our proposition fared better in terms of accuracy and execution time. [ABSTRACT FROM AUTHOR]

Social media provide plenty of textual information in various languages. This information can contain or provoke hatred towards different social or religious groups. In this paper, we study methods to process short text messages in English, Hindi, and Russian and identify such intolerance with cross-lingual Transformer models. Moreover, these models can be easily adapted to analyze other languages. We fine-tuned these models with several training techniques to build accurate hate speech detectors that are robust to adversarial attacks. Additional preprocessing was carried out for all datasets to improve the quality of model training. Also, for one of the training datasets, we applied the text attack algorithm that replaces some words with synonyms. For some languages, such an attack can greatly reduce the quality of the model. Experiment results show that mixing adversarial examples to a training dataset and combining deep models to randomized ensembles allows not only to reduce test error on attacked data for languages from the dataset (Hindi, Russian) but also to achieve better accuracy in other languages. [ABSTRACT FROM AUTHOR]

In the face of uncontrolled offensive content on social media, automated detection emerges as a critical need. This paper tackles this challenge by proposing a novel approach for identifying offensive language in multilingual, code-mixed, and script-mixed settings. The study presents a novel multilingual hybrid dataset constructed by merging diverse monolingual and bilingual resources. Further, we systematically evaluate the impact of input representations (Word2Vec, Global Vectors for Word Representation (or GloVe), Bidirectional Encoder Representations from Transformers (or BERT), and uniform initialization) and deep learning models (Convolutional Neural Network (or CNN), Bidirectional Long Short Term Memory (or Bi-LSTM), Bi-LSTM-Attention, and fine-tuned BERT) on detection accuracy. Our comprehensive experiments on a dataset of 42,560 social media comments from five languages (English, Hindi, German, Tamil, and Malayalam) reveal the superiority of fine-tuned BERT. Notably, it achieves a macro average F 1 -score of 0.79 for monolingual tasks and an impressive 0.86 for code-mixed and script-mixed tasks. These findings significantly advance offensive language detection methodologies and shed light on the complex dynamics of multilingual social media, paving the way for more inclusive and safer online communities. [ABSTRACT FROM AUTHOR]

Over the past decade, increased use of social media has led to an increase in hate content. To address this issue new solutions must be implemented to filter out this kind of inappropriate content. Because manual filtering is difficult, several studies have been conducted in order to automate the process. This paper introduces a method based on a combination of Glove and FastText word embedding as input features and a BiGRU model to identify hate speech from social media websites. The obtained results show that our proposed model (BiGRU Glove FT) is effective in detecting inappropriate content. This model detect hate speech on OLID dataset, using an effective learning process that classifies the text into offensive and not offensive language. The performance of the system attained 84%, 87%, 93%, 90% accuracy, precision, recall, and f1-score respectively. [ABSTRACT FROM AUTHOR]

Nowadays, users across the globe interact with one another for information exchange, communication, and association on various online social media. However, some individuals exploit these venues for malicious practices like hate speech and cyberbully. In this paper, we present an improved multilingual hate speech and cyberbully detection model using bagging-stacking based hybrid ensemble deep learning techniques. The proposed model utilizes Bi-directional Long Short-Term Memory (BiLSTM), Bi-directional Gated Recurrent Unit (Bi-GRU), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM) techniques to enhance the overall performance. We first preprocess the multilingual data streams followed by adoption of Global vectors for word Representation (GloVe) embeddings to convert words to a vector representation in parallel enabling the data streams for binary classification task. In order to construct an architecture for the detection of hate speech and cyberbully, we introduce a heterogeneous fusion of multiple effective models in a unique approach such that CNN-LSTM utilizes a stacking approach with stochastic gradient descent to achieve optimal weights, whereas all the base learners used bagging ensemble approach with cross-validation to reach optimal weights. The final output layer of the proposed ensemble deep learning architecture is achieved using a super learner approach on base learners. To show the efficacy of the proposed model, we conduct the simulation on a total of nine real-world social media datasets in different languages and compared the results with other contemporary hate speech and cyberbully detection methods. The collected findings show that the proposed model outperforms other models on considered datasets and shows an improvement of at least 4.44% in F1 scores. • Proposed an ensemble DL multilingual model for hate speech & cyberbully detection. • Model combines bagging, stacking, and super learning methods to improved results. • Model adopts a light weight heterogeneous fusion and bagging-stacking hybrid methods. • The experimental findings advocate the superiority of the proposed model. [ABSTRACT FROM AUTHOR]

Along with the COVID-19 pandemic, an "infodemic" of false and misleading information has emerged and has complicated the COVID-19 response efforts. Social networking sites such as Facebook and Twitter have contributed largely to the spread of rumors, conspiracy theories, hate, xenophobia, racism, and prejudice. To combat the spread of fake news, researchers around the world have and are still making considerable efforts to build and share COVID-19 related research articles, models, and datasets. This paper releases "AraCOVID19-MFH"1 a manually annotated multi-label Arabic COVID-19 fake news and hate speech detection dataset. Our dataset contains 10,828 Arabic tweets annotated with 10 different labels. The labels have been designed to consider some aspects relevant to the fact-checking task, such as the tweet's check worthiness, positivity/negativity, and factuality. To confirm our annotated dataset's practical utility, we used it to train and evaluate several classification models and reported the obtained results. Though the dataset is mainly designed for fake news detection, it can also be used for hate speech detection, opinion/news classification, dialect identification, and many other tasks. [ABSTRACT FROM AUTHOR]

During the COVID-19 pandemic, online social media platforms such as Twitter facilitate the exchange of information among people. However, the prevalence of "infodemic" such as online hate speech has exacerbated social rifts, discrimination, prejudice and even hate crimes. Timely and effective detection of the hate speech will help create a healthy public opinion environment. Most of the current COVID-19-related hate speech research focuses on a single language, such as English. In this paper, we introduce a cross-lingual transfer learning method, aiming to contribute to hate speech detection in low-resource languages. We propose a deep learning based model to classify hate speech with a pre-trained language model for multilingual text embedding. Data augmentation and cross-lingual contrastive learning are then utilized to further improve the performance of cross-lingual knowledge transfer. To evaluate our method, we collected three publicly available annotated COVID-19-related hate speech datasets on Twitter, i.e., two in English and one in German. Furthermore, a Chinese dataset based on Weibo is constructed to expand multilingual data. The experimental results across three languages illustrate the effectiveness of our method for cross-lingual hate speech detection. Test F1-scores of our method for English, Chinese, German as transfer target languages can reach up to 0.728, 0.799 and 0.612 respectively, which are on average better than other baselines. • Suggested a cross-lingual COVID-19-related hate speech detection problem. • Introduced cross-lingual contrastive learning. • Validated model performance on datasets in three languages. [ABSTRACT FROM AUTHOR]

People increasingly use social media platforms to express themselves by posting visuals and texts. As a result, hate content is on the rise, necessitating practical visual caption analysis. Thus, the relationship between image and caption modalities is crucial in visual caption analysis. Contrarily, most methods combine features from the image and caption modalities using deep learning architectures with millions of parameters already trained without integrating a specialized attention module, resulting in less desirable outcomes. This paper suggests a novel multi-modal architecture for identifying hateful memetic information in response to the above observation. The proposed architecture contains a novel "multi-scale kernel attentive visual" (MSKAV) module that uses an efficient multi-branch structure to extract discriminative visual features. Additionally, MSKAV utilizes an adaptive receptive field using multi-scale kernels. MSKAV also incorporates a multi-directional visual attention module to highlight spatial regions of importance. The proposed model also contains a novel "knowledge distillation-based attentional caption" (KDAC) module. It uses a transformer-based self-attentive block to extract discriminative features from meme captions. Thorough experimentation on multi-modal hate speech benchmarks MultiOff, Hateful Memes, and MMHS150K datasets achieved accuracy scores of 0.6250, 0.8750, and 0.8078, respectively. It also reaches impressive AUC scores of 0.6557, 0.8363, and 0.7665 on the three datasets, respectively, beating SOTA multi-modal hate speech identification models. [ABSTRACT FROM AUTHOR]

Social media platforms have become an essential means of communication, but they also serve as a breeding ground for hateful content. Detecting hate speech accurately is challenging due to factors such as slang and implicit hate speech. In response to these challenges, this paper presents a novel ensemble approach utilizing DeBERTa models, integrating back-translation and GPT-3 augmentation techniques during both training and test time. This method aims to address the complexities associated with detecting hate speech, resulting in more robust and accurate results. Our findings indicate that the proposed approach significantly enhances hate speech detection performance across various metrics and models in both the Parler and GAB datasets. For reproducibility and further exploration, our code is publicly available at https://github.com/OrKatz7/parler-hate-speech. • Apply Test-Time Augmentation in NLP to improve hate speech detection. • Enhance augmentation quality with a selection method to filter unsuitable ones. • Merge DeBERTa, back-translation & GPT-3 to boost hate speech detection. • Manage hate speech complexities - effectively detects slang & implicit language. [ABSTRACT FROM AUTHOR]

Social media platforms have evolved into an online representation of our social interactions. We may use the resources they provide to analyze phenomena that occur within them, such as the development and viralization of offensive and hostile content. In today's polarized world, the escalating nature of this behavior is cause for concern in modern society. This research includes an in-depth examination of previous efforts and strategies for detecting and preventing hateful content on the social network Twitter, as well as a novel classification approach based on users' profiles, related social environment and generated tweets. This paper's contribution is threefold: (i) an improvement in the performance of the HaterNet algorithm, an expert system developed in collaboration with the Spanish National Office Against Hate Crimes of the Spanish State Secretariat for Security (Ministry of the Interior) that is capable of identifying and monitoring the evolution of hate speech on Twitter using an LTSM + MLP neural network architecture. To that end, a model based on BERT, HaterBERT , has been created and tested using HaterNet 's public dataset, providing results that show a significant improvement; (ii) A methodology to create a user database in the form of a relational network to infer textual and centrality features. This contribution, SocialGraph , has been independently tested with various traditional Machine Learning and Deep Learning algorithms, demonstrating its usefulness in spotting haters; (iii) a final model, SocialHaterBERT , that integrates the previous two approaches by analyzing features other than those inherent in the text. Experiment results reveal that this last contribution greatly improves outcomes, establishing a new field of study that transcends textual boundaries, paving the way for future research in coupled models from a diachronic and dynamic perspective. • HaterBERT, a BETO-based Spanish hate speech classifier that outperforms previous. • SocialGraph, a methodology to create a database representing user's social context. • SocialHaterBERT, a novel multimodal model that combines HaterBERT and SocialGraph. • Demonstration of how the context of social media improves textual classification. [ABSTRACT FROM AUTHOR]

• The first public dataset dedicated to detecting abusive Turkish messages. • 10,528 abusive, 19,826 not-abusive Instagram comments have been collected. • CNN, NB, SVM, DT, RF, LR, AdaBoost, and XGBoost classifiers were evaluated. • The best performance (F1-score: 0.974) was achieved by the CNN model. Instagram is a free photo-sharing platform where each user has a profile and can upload photos for followers to view, like, and comment. Abusive comments on images can be humiliating and harmful to those who share photos. Developing a comment filter in languages other than English is difficult and time-consuming. This paper proposes a dataset called Abusive Turkish Comments (ATC) to detect abusive Instagram comments in Turkish. It is composed of a large number of Instagram comments posted to tabloid and sports accounts (i.e., 10,528 abusive and 19,826 not-abusive). It is the first public dataset dedicated to detecting abusive Turkish messages, as far as we know. The sentiment annotation has been done in sentence-level by assigning polarity to each comment. The performance of the abusive message detection models was evaluated using several performance metrics: Convolutional Neural Network (CNN), five well-known classifiers (i.e., Naive Bayes, Support Vector Machine, Decision Tree, Random Forest, and Logistic Regression), and two reweighted classifiers (i.e., Adaptive Boosting (AdaBoost), eXtreme Gradient Boosting (XGBoost)) were compared in terms of F1-score, precision, and recall. The results showed that the best performance (i.e., Micro-averaged F1-score: 0.974, Macro-averaged F1-score: 0.973, Kappa-value: 0.946) was yielded by the CNN model on the oversampled ATC dataset. The abusive message detection model proposed in this study can contribute to the development of Turkish comment filters on Instagram. Different model combinations are considered to select the best model that gives better recognition accuracy. [ABSTRACT FROM AUTHOR]

Nowadays, due to the great uncontrolled content posted daily on the Web, there has also been a huge increase in the dissemination of hate speech worldwide. Social media, blogs and community forums are examples where people are freely allowed to communicate. However, freedom of expression is not always respectful since offensive or insulting language is sometimes used. Social media companies often rely on users and content moderators to report on this type of content. Nevertheless, due to the large amount of content generated every day on the Web, automatic systems based on Natural Language Processing techniques are required for identifying abusive language online. To date, most of the systems developed to combat this problem are mainly focused on English content, but this issue is a worldwide concern and therefore other languages such as Spanish are involved. In this paper, we address the task of Spanish hate speech identification on social media and provide a deeper understanding of the capabilities of new techniques based on machine learning. In particular, we compare the performance of Deep Learning methods with recently pre-trained language models based on Transfer Learning as well as with traditional machine learning models. Our main contribution is the achievement of promising results in Spanish by applying multilingual and monolingual pre-trained language models such as BERT, XLM and BETO. • Addressing hate speech detection on social media for Spanish. • Comparing the performance of Deep Learning and Machine Learning models. • Evaluating recent pre-trained language models based on Transformer mechanism. • Comparing pre-trained multilingual and monolingual models available for Spanish. • Conducting an error analysis in order to understand the difficulty of the task. [ABSTRACT FROM AUTHOR]

• Developed browser plugins to visualize hate speech live on Facebook & Twitter. • Designed proof of concepts for personalized dashboards for Twitter & Instagram. • Optimized classifiers for Hate Speech with tests on standard datasets. • Application areas for the tool are discussed. The multi-fold growth of the social media user-base fuelled a substantial increase in the amount of hate speech posts on social media platforms. The enormous data volume makes it hard to capture such cases and either moderate or delete them. This paper presents an approach to detect and visualize online aggression, a special case of hate speech, over social media. Aggression is categorized into overtly aggressive (OAG), covertly aggressive (CAG), and non-aggressive labels (NAG). We have designed a user interface based on a web browser plugin over Facebook and Twitter to visualize the aggressive comments posted on the Social media user's timelines. This plugin interface might help to the security agency to keep a tab on the social media stream. It also provides citizens with a tool that is typically only available for large enterprises. The availability of such a tool alleviates the technological imbalance between industry and citizens. Besides, the system might be helpful to the research community to create further tools and prepare weakly labeled training data in a few minutes using comments posted by users on celebrity's Facebook, Twitter timeline. We have reported the results on a newly created dataset of user comments posted on Facebook and Twitter using our proposed plugins and the standard Trolling Aggression Cyberbullying 2018 (TRAC) dataset in English and code-mixed Hindi. Various classifiers like Support Vector Machine (SVM), Logistic regression, deep learning model based on Convolution Neural Network (CNN), Attention-based model, and the recently proposed BERT pre-trained language model by Google AI, have been used for aggression classification. The weighted F1-score of around 0.64 and 0.62 is achieved on TRAC Facebook English and Hindi datasets while on Twitter English and Hindi datasets, the weighted F1-score is 0.58 and 0.50, respectively. [ABSTRACT FROM AUTHOR]