Search results

Hate speech is an expression of intense hatred. Twitter has become a popular analytical tool for the prediction and monitoring of abusive behaviors. Hate speech detection with social media data has witnessed special research attention in recent studies, hence, the need to design a generic metadata architecture and efficient feature extraction technique to enhance hate speech detection.

This study proposes a hybrid embeddings enhanced with a topic inference method and an improved cuckoo search neural network for hate speech detection in Twitter data. The proposed method uses a hybrid embeddings technique that includes Term Frequency-Inverse Document Frequency (TF-IDF) for word-level feature extraction and Long Short Term Memory (LSTM) which is a variant of recurrent neural networks architecture for sentence-level feature extraction. The extracted features from the hybrid embeddings then serve as input into the improved cuckoo search neural network for the prediction of a tweet as hate speech, offensive language or neither.

The proposed method showed better results when tested on the collected Twitter datasets compared to other related methods. In order to validate the performances of the proposed method, t-test and post hoc multiple comparisons were used to compare the significance and means of the proposed method with other related methods for hate speech detection. Furthermore, Paired Sample t-Test was also conducted to validate the performances of the proposed method with other related methods.

Finally, the evaluation results showed that the proposed method outperforms other related methods with mean F1-score of 91.3.

The main novelty of this study is the use of an automatic topic spotting measure based on naïve Bayes model to improve features representation.

This paper aims to propose an approach for hate speech detection against politicians in Arabic community on social media (e.g. Youtube). In the literature, similar works have been presented for other languages such as English. However, to the best of the authors’ knowledge, not much work has been conducted in the Arabic language.

This approach uses both classical algorithms of classification and deep learning algorithms. For the classical algorithms, the authors use Gaussian NB (GNB), Logistic Regression (LR), Random Forest (RF), SGD Classifier (SGD) and Linear SVC (LSVC). For the deep learning classification, four different algorithms (convolutional neural network (CNN), multilayer perceptron (MLP), long- or short-term memory (LSTM) and bi-directional long- or short-term memory (Bi-LSTM) are applied. For extracting features, the authors use both Word2vec and FastText with their two implementations, namely, Skip Gram (SG) and Continuous Bag of Word (CBOW).

Simulation results demonstrate the best performance of LSVC, BiLSTM and MLP achieving an accuracy up to 91%, when it is associated to SG model. The results are also shown that the classification that has been done on balanced corpus are more accurate than those done on unbalanced corpus.

The principal originality of this paper is to construct a new hate speech corpus (Arabic_fr_en) which was annotated by three different annotators. This corpus contains the three languages used by Arabic people being Arabic, French and English. For Arabic, the corpus contains both script Arabic and Arabizi (i.e. Arabic words written with Latin letters). Another originality is to rely on both shallow and deep leaning classification by using different model for extraction features such as Word2vec and FastText with their two implementation SG and CBOW.

The present study was designed to investigate eight research questions that are related to the analysis and the detection of dialectal Arabic hate speech that targeted African refugees and illegal migrants on the YouTube Algerian space.

The transfer learning approach which recently presents the state-of-the-art approach in natural language processing tasks has been exploited to classify and detect hate speech in Algerian dialectal Arabic. Besides, a descriptive analysis has been conducted to answer the analytical research questions that aim at measuring and evaluating the presence of the anti-refugee/migrant discourse on the YouTube social platform.

Data analysis revealed that there has been a gradual modest increase in the number of anti-refugee/migrant hateful comments on YouTube since 2014, a sharp rise in 2017 and a sharp decline in later years until 2021. Furthermore, our findings stemming from classifying hate content using multilingual and monolingual pre-trained language transformers demonstrate a good performance of the AraBERT monolingual transformer in comparison with the monodialectal transformer DziriBERT and the cross-lingual transformers mBERT and XLM-R.

Automatic hate speech detection in languages other than English is quite a challenging task that the literature has tried to address by various approaches of machine learning. Although the recent approach of cross-lingual transfer learning offers a promising solution, tackling this problem in the context of the Arabic language, particularly dialectal Arabic makes it even more challenging. Our results cast a new light on the actual ability of the transfer learning approach to deal with low-resource languages that widely differ from high-resource languages as well as other Latin-based, low-resource languages.

Hate speech in recent times has become a troubling development. It has different meanings to different people in different cultures. The anonymity and ubiquity of the social media provides a breeding ground for hate speech and makes combating it seems like a lost battle. However, what may constitute a hate speech in a cultural or religious neutral society may not be perceived as such in a polarized multi-cultural and multi-religious society like Nigeria. Defining hate speech, therefore, may be contextual. Hate speech in Nigeria may be perceived along ethnic, religious and political boundaries. The purpose of this paper is to check for the presence of hate speech in social media platforms like Twitter, and to what degree is hate speech permissible, if available? It also intends to find out what monitoring mechanisms the social media platforms like Facebook and Twitter have put in place to combat hate speech. Lexalytics is a term coined by the authors from the words lexical analytics for the purpose of opinion mining unstructured texts like tweets.

This research developed a Python software called polarized opinions sentiment analyzer (POSA), adopting an ego social network analytics technique in which an individual’s behavior is mined and described. POSA uses a customized Python N-Gram dictionary of local context-based terms that may be considered as hate terms. It then applied the Twitter API to stream tweets from popular and trending Nigerian Twitter handles in politics, ethnicity, religion, social activism, racism, etc., and filtered the tweets against the custom dictionary using unsupervised classification of the texts as either positive or negative sentiments. The outcome is visualized using tables, pie charts and word clouds. A similar implementation was also carried out using R-Studio codes and both results are compared and a t-test was applied to determine if there was a significant difference in the results. The research methodology can be classified as both qualitative and quantitative. Qualitative in terms of data classification, and quantitative in terms of being able to identify the results as either negative or positive from the computation of text to vector.

The findings from two sets of experiments on POSA and R are as follows: in the first experiment, the POSA software found that the Twitter handles analyzed contained between 33 and 55 percent hate contents, while the R results show hate contents ranging from 38 to 62 percent. Performing a t-test on both positive and negative scores for both POSA and R-studio, results reveal p-values of 0.389 and 0.289, respectively, on an α value of 0.05, implying that there is no significant difference in the results from POSA and R. During the second experiment performed on 11 local handles with 1,207 tweets, the authors deduce as follows: that the percentage of hate contents classified by POSA is 40 percent, while the percentage of hate contents classified by R is 51 percent. That the accuracy of hate speech classification predicted by POSA is 87 percent, while free speech is 86 percent. And the accuracy of hate speech classification predicted by R is 65 percent, while free speech is 74 percent. This study reveals that neither Twitter nor Facebook has an automated monitoring system for hate speech, and no benchmark is set to decide the level of hate contents allowed in a text. The monitoring is rather done by humans whose assessment is usually subjective and sometimes inconsistent.

This study establishes the fact that hate speech is on the increase on social media. It also shows that hate mongers can actually be pinned down, with the contents of their messages. The POSA system can be used as a plug-in by Twitter to detect and stop hate speech on its platform. The study was limited to public Twitter handles only. N-grams are effective features for word-sense disambiguation, but when using N-grams, the feature vector could take on enormous proportions and in turn increasing sparsity of the feature vectors.

The findings of this study show that if urgent measures are not taken to combat hate speech there could be dare consequences, especially in highly polarized societies that are always heated up along religious and ethnic sentiments. On daily basis tempers are flaring in the social media over comments made by participants. This study has also demonstrated that it is possible to implement a technology that can track and terminate hate speech in a micro-blog like Twitter. This can also be extended to other social media platforms.

This study will help to promote a more positive society, ensuring the social media is positively utilized to the benefit of mankind.

The findings can be used by social media companies to monitor user behaviors, and pin hate crimes to specific persons. Governments and law enforcement bodies can also use the POSA application to track down hate peddlers.

This study aims to examine the role of hope and hate in political leaders’ messages in influencing liberals versus conservatives’ social-distancing behavior during the COVID-19 pandemic. Given the increasing political partisanship across the world today, using the appropriate message framing has important implications for social and public policy.

The authors use two Natural Language Processing (NLP) methods – a pretrained package (HateSonar) and a classifier built to implement our supervised neural network-based model architecture using RoBERTa – to analyze 61,466 tweets by each US state’s governor and two senators with the goal of examining the association between message factors invoking hate and hope and increased or decreased social distancing from March to May 2020. The authors examine individuals’ social-distancing behaviors (the amount of nonessential driving undertaken) using data from 3,047 US counties between March 13 and May 31, 2020, as reported by Google COVID-19 Community Mobility Reports and the New York Times repository of COVID-19 data.

The results show that for conservative state leaders, the use of hate increases nonessential driving of state residents. However, when these leaders use hope in their speech, nonessential driving of state residents decreases. For liberal state leaders, the use of hate displays a directionally different result as compared to their conservative counterparts.

Amid the emergence of new analytic techniques and novel data sources, the findings demonstrate that the use of global positioning systems data and social media analysis can provide valuable and precise insights into individual behavior. They also contribute to the literature on political ideology and emotion by demonstrating the use of specific emotion appeals in targeting specific consumer segments based on their political ideology.

The findings have significant implications for policymakers and public health officials regarding the importance of considering partisanship when developing and implementing public health policies. As partisanship continues to increase, applying the appropriate emotion appeal in messages will become increasingly crucial. The findings can help marketers and policymakers develop more effective social marketing campaigns by tailoring specific appeals given the political identity of the consumer.

Using Neural NLP methods, this study identifies the specific factors linking social media messaging from political leaders and increased compliance with health directives in a partisan population.

Online hate speech (OHS) is becoming a serious problem worldwide including in Bangladesh. This study aims to understand the nature of OHS against religious groups and explore its impact on their social life and mental health.

A qualitative approach was followed and 11 in-depth interviews (IDIs) were conducted with the selected OHS victims. This study conducted a semi-structural interview using Google Form following the design questionnaire for selecting IDIs participants.

This study found that religious minorities experience online hatred through online media by the major religious group in Bangladesh. Natures of OHS are commenting on social media posts, sharing hateful memes and sending private messages using slang language targeting religious identity, religious food habits and ethnic identities. Victims were offended, abused and bullied by unknown persons, their university friends and colleagues. Victims of OHS did not take any action against it due to fear of insecurity. Victims of OHS felt low-minded, helpless and anxious after the experience of OHS; they felt more insecure and vulnerable socially and mentally.

The findings of this study suggest that policymakers identify the nature of OHS and take proper steps for reducing the frequency of OHS in Bangladesh. To combat the OHS, authorities have to make legal enforcement equal for everyone.

Social media (SM) is a new communication tool that substantially contribute to facilitating online hate speech (OHS). In emphasis of the question “what role can SM play in an individual’s life?”, this study aims to understand Bangladeshi university students’ personal experiences and opinions of OHSs related to SM.

The authors used an online survey method to collect data and retrieved responses from 410 students. Mann–Whitney U test, Kruskal–Wallis test and Spearman’s rank correlation analysis were used to test the hypotheses.

This study found that hate speech is a familiar term among students. Students’ political views or opinions, religion and gender have become the most targeted instruments for OHSs. Comparing students’ use of SM, the authors found that Facebook was the most used SM site to spread hate speech in Bangladesh. In terms of personal experiences, the findings indicated that 45.6% of students became victims of OHSs at least once or more times, and the majority of students tended to simply avoid OHSs. Another significant finding was that OHS has real-life effects on the students, resulting in various personal and psychological distress.

Although some research has been conducted on hate speech at the local level, to the best of the authors’ knowledge, no study has focused on the student community. To the best of the authors’ knowledge, this study is the first attempt in Bangladesh to focus on OHSs from a student’s personal viewpoint.

The purpose of this paper is to present a quantitative and qualitative synthesis of the diverse academic proposals and initiatives for preventing and eliminating hate speech on the internet.

The foundation for this study is a systematic review of papers devoted to the analysis of hate speech. It has been conducted using the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) protocol and applied to an initial corpus of 436 academic texts. Having implemented the suitability, screening and inclusion criteria, this corpus was refined to a sample of 74 articles.

The main subject categories studied in this corpus of academic research are legal issues and social media. In the majority of the articles, the use of hate speech via social media is associated with five typologies: religion, cyber racism, political slurs, misogyny and attacks on the LGTBI community. The absence of ethical reflection is one of the major shortcomings of IT-focused research and analysis devoted to online hate speech.

To date various systematic reviews have been presented, and they focus on detecting or describing hate speech. These have used either the search appraisal synthesis analysis framework or the Cochrane network. The PRISMA protocol was applied for this study, and both Scopus and texts in German were included. To date no major, rigorous systematic review has been undertaken of proposals to combat hate speech.

The link between hate speech and poverty has not been studied in depth within the academic sphere. Tolerance and ethical compassion are not granted the attention they merit when it comes to analysing the phenomenon of hate speech.

The purpose of this paper is to examine strengths and limitations that search engines (SEs) exhibit when responding to web search queries associated with the grade school curriculum

The authors employed a simulation-based experimental approach to conduct an in-depth empirical examination of SEs and used web search queries that capture information needs in different search scenarios.

Outcomes from this study highlight that child-oriented SEs are more effective than traditional ones when filtering inappropriate resources, but often fail to retrieve educational materials. All SEs examined offered resources at reading levels higher than that of the target audience and often prioritized resources with popular top-level domain (e.g. “.com”).

Findings have implications for human intervention, search literacy in schools, and the enhancement of existing SEs. Results shed light on the impact on children’s education that result from introducing misconception about SEs when these tools either retrieve no results or offer irrelevant resources, in response to web search queries pertinent to the grade school curriculum.

The authors examined child-oriented and popular SEs retrieval of resources aligning with task objectives and user capabilities–resources that match user reading skills, do not contain hate-speech and sexually-explicit content, are non-opinionated, and are curriculum-relevant. Findings identified limitations of existing SEs (both directly or indirectly supporting young users) and demonstrate the need to improve SE filtering and ranking algorithms.

Expressional anomie (e.g. obscene words) can hinder communications and even obstruct improvements of national literacy. Meanwhile, the borderless and rapid transmission of the internet has exacerbated the influences. Hence, the purpose of this paper is detecting online anomic expression automatically and analyzing dynamic evolution processes of expressional anomie, so as to reveal multidimensional status of expressional anomie.

This paper conducted expressional anomie analysis via fine-grained microblog mining. Specifically, anomic microblogs and their anomic types were identified via a supervised classification method. Then, the evolutions of expressional anomie were analyzed, and impacts of users’ characteristics on the evolution process were mined. Finally, expressional anomie characteristics and evolution trends were obtained.

Empirical results on microblogs indicate that more effective and diversified measures need to be used to address the current large-scale anomie in expression. Moreover, measures should be tailored to individuals and local conditions.

To the best of the authors’ knowledge, it is the first research to mine evolutions of expressional anomie automatically in social media. It may discover more continuous and universal rules of expressional anomie, so as to optimize the online expression environment.