Automatic cyberbullying detection: A systematic review

A novel approach toward cyberbullying with intelligent recommendations using deep learning based blockchain solution

Integrating healthcare into traffic accident prevention through predictive modeling holds immense potential. Decentralized Defense presents a transformative vision for combating cyberbullying, prioritizing user privacy, fostering a safer online environment, and offering valuable insights for both healthcare and predictive modeling applications. As cyberbullying proliferates in social media, a pressing need exists for a robust and innovative solution that ensures user safety in the cyberspace. This paper aims toward introducing the approach of merging Blockchain and Federated Learning (FL), to create a decentralized AI solutions for cyberbullying. It has also used Alloy Language for formal modeling of social connections using specific declarations that are defined by the novel algorithm in the paper on two different datasets on Cyberbullying and are available online. The proposed novel method uses DBN to run established relation tests amongst the features in two phases, the first is LSTM to run tests to develop established features for the DBN layer and second is that these are run on various blocks of information of the blockchain. The performance of our proposed research is compared with the previous research and are evaluated using several metrics on creating the standard benchmarks for real world applications.

Digital Disconnection: A Qualitative Study of Youth and Young Adult Perspectives on Cyberbullying and the Adoption of Auto-Detection or Software Tools

PURPOSE

The purpose of this study was to understand the needs of youth and young adults, current gaps around safeguarding social media, and factors affecting adoption of data-driven auto-detection or software tools.

METHODS

This qualitative study is the first step of a larger initiative that aims to use participatory action research and co-design principles to develop a digital tool that targets cyberbullying. Youth and young adults aged 16-21 years were recruited to participate in semistructured focus groups between March 2020 and November 2021. Thematic analysis was used to develop themes, with a member-checking process to validate the findings.

RESULTS

Six focus groups were completed with 39 participants and five themes were generated from the analysis. Participants described the mental health impacts of cyberbullying on young people, the stigma associated with it, and the need for more mental health resources. They felt that additional efforts are needed to improve the school environment, school-based interventions, and training protocols to ensure that youth feel safe reporting cyberbullying. Most participants were open to using a digital solution but raised concerns around the trustworthiness of artificial intelligence and wanted it to be co-designed with young people, integrated across platforms, informed by data-driven decisions, and transparent with users.

DISCUSSION

Youth and young adults are accepting of a low-risk digital cyberbullying solution as current interventions are not meeting their needs.

Site Agnostic Approach to Early Detection of Cyberbullying on Social Media Networks

The rise in the use of social media networks has increased the prevalence of cyberbullying, and time is paramount to reduce the negative effects that derive from those behaviours on any social media platform. This paper aims to study the early detection problem from a general perspective by carrying out experiments over two independent datasets (Instagram and Vine), exclusively using users' comments. We used textual information from comments over baseline early detection models (fixed, threshold, and dual models) to apply three different methods of improving early detection. First, we evaluated the performance of Doc2Vec features. Finally, we also presented multiple instance learning (MIL) on early detection models and we assessed its performance. We applied timeawareprecision (TaP) as an early detection metric to asses the performance of the presented methods. We conclude that the inclusion of Doc2Vec features improves the performance of baseline early detection models by up to 79.6%. Moreover, multiple instance learning shows an important positive effect for the Vine dataset, where smaller post sizes and less use of the English language are present, with a further improvement of up to 13%, but no significant enhancement is shown for the Instagram dataset.

A Review on Deep-Learning-Based Cyberbullying Detection

Bullying is described as an undesirable behavior by others that harms an individual physically, mentally, or socially. Cyberbullying is a virtual form (e.g., textual or image) of bullying or harassment, also known as online bullying. Cyberbullying detection is a pressing need in today’s world, as the prevalence of cyberbullying is continually growing, resulting in mental health issues. Conventional machine learning models were previously used to identify cyberbullying. However, current research demonstrates that deep learning surpasses traditional machine learning algorithms in identifying cyberbullying for several reasons, including handling extensive data, efficiently classifying text and images, extracting features automatically through hidden layers, and many others. This paper reviews the existing surveys and identifies the gaps in those studies. We also present a deep-learning-based defense ecosystem for cyberbullying detection, including data representation techniques and different deep-learning-based models and frameworks. We have critically analyzed the existing DL-based cyberbullying detection techniques and identified their significant contributions and the future research directions they have presented. We have also summarized the datasets being used, including the DL architecture being used and the tasks that are accomplished for each dataset. Finally, several challenges faced by the existing researchers and the open issues to be addressed in the future have been presented.

A Systematic Literature Review of the Use of Computational Text Analysis Methods in Intimate Partner Violence Research

Purpose

Computational text mining methods are proposed as a useful methodological innovation in Intimate Partner Violence (IPV) research. Text mining can offer researchers access to existing or new datasets, sourced from social media or from IPV-related organisations, that would be too large to analyse manually. This article aims to give an overview of current work applying text mining methodologies in the study of IPV, as a starting point for researchers wanting to use such methods in their own work.

Methods

This article reports the results of a systematic review of academic research using computational text mining to research IPV. A review protocol was developed according to PRISMA guidelines, and a literature search of 8 databases was conducted, identifying 22 unique studies that were included in the review.

Results

The included studies cover a wide range of methodologies and outcomes. Supervised and unsupervised approaches are represented, including rule-based classification (n = 3), traditional Machine Learning (n = 8), Deep Learning (n = 6) and topic modelling (n = 4) methods. Datasets are mostly sourced from social media (n = 15), with other data being sourced from police forces (n = 3), health or social care providers (n = 3), or litigation texts (n = 1). Evaluation methods mostly used a held-out, labelled test set, or k-fold Cross Validation, with Accuracy and F1 metrics reported. Only a few studies commented on the ethics of computational IPV research.

Conclusions

Text mining methodologies offer promising data collection and analysis techniques for IPV research. Future work in this space must consider ethical implications of computational approaches.

Assessing Cyberbullying in Adolescence: New Evidence for the Spanish Version of the European Cyberbullying Intervention Project Questionnaire (ECIP-Q)

The prevention of cyberbullying at school requires assessing its prevalence by means of brief measurement instruments with adequate psychometric properties. The present study aims to study the psychometric properties of the European Cyberbullying Intervention Project Questionnaire (ECIP-Q) in a sample of 1777 Spanish adolescents (54.1% women, M = 15.71 years; SD = 1.26), selected by stratified random cluster sampling. The two-factor model (victimization and aggression) displayed appropriate goodness of-fit indices. Configural measurement invariance model across gender was found. The omega reliability coefficient for the victimization subscale was 0.82, and for the aggression subscale was 0.68. The ECIP-Q scores were negatively associated with self-esteem and prosocial behavior, and positively associated with depression symptoms and emotional and behavioral difficulties. Significant differences were found between victim and non-victim groups, and between aggressor and non-aggressor groups on the same variables. Victims and aggressors scored lower on self-esteem, and higher on depression symptoms and emotional and behavioral difficulties than those not involved in cyberbullying situations. These findings contribute to demonstrate the satisfactory psychometric quality of the ECIP-Q scores as an assessment tool for cyberbullying in Spanish adolescents.

Understanding Bullying and Cyberbullying Through an Ecological Systems Framework: the Value of Qualitative Interviewing in a Mixed Methods Approach

Recognized as complex and relational, researchers endorse a systems/social-ecological framework in examining bullying and cyberbullying. According to this framework, bullying and cyberbullying are examined across the nested social contexts in which youth live—encompassing individual features; relationships including family, peers, and educators; and ecological conditions such as digital technology. Qualitative inquiry of bullying and cyberbullying provides a research methodology capable of bringing to the fore salient discourses such as dominant social norms and otherwise invisible nuances such as motivations and dilemmas, which might not be accessed through quantitative studies. Through use of a longitudinal and multi-perspective mixed methods study, the purpose of the current paper is to demonstrate the ways qualitative interviews contextualize quantitative findings and to present novel discussion of how qualitative interviews explain and enrich the quantitative findings. The following thematic areas emerged and are discussed: augmenting quantitative findings through qualitative interviews, contextualizing new or rapidly evolving areas of research, capturing nuances and complexity of perspectives, and providing moments for self-reflection and opportunities for learning.

A critical review of the Online Safety Bill

The UK Parliament has tabled the Online Safety Bill to make the internet safer for users by requiring providers to regulate legal but harmful content on their platform. This paper critically assesses the draft legislation, surveying its rationale; its scope in terms of lawful and unlawful harms it intends to regulate; and the mechanisms through which it will be enforced. We argue that it requires further refinement if it is to protect free speech and innovation in the digital sphere. We propose four conclusions: further evidence is required to substantiate the necessity and proportionality of the Bill’s interventions; the Bill risks a democratic deficit by limiting the opportunity for parliamentary scrutiny; the duties of the bill may be too wide (in terms of burdening providers); and that enforcement of a Code of Practice will likely be insufficient.

This critical perspective makes a timely contribution to the tech policy debate concerning the monitoring and moderation of online content. Governments globally are currently considering a range of legislative interventions to limit online abuse, disinformation, and the dissemination of illegal content on social media platforms. These interventions will significantly impact online free speech, competition between platforms, and the democratic function of online platforms. By investigating the UK’s Online Safety Bill, comparing it with similar interventions, and considering the political impact of different digital tools for moderation, this perspective aims to inform the current policy debate by combining technical and political insight. It indicates the need for further research into the comparative efficacy of different methods of content monitoring and moderation.

Building towards Automated Cyberbullying Detection: A Comparative Analysis

The increased use of social media among digitally anonymous users, sharing their thoughts and opinions, can facilitate participation and collaboration. However, this anonymity feature which gives users freedom of speech and allows them to conduct activities without being judged by others can also encourage cyberbullying and hate speech. Predators can hide their identity and reach a wide range of audience anytime and anywhere. According to the detrimental effect of cyberbullying, there is a growing need for cyberbullying detection approaches. In this survey paper, a comparative analysis of the automated cyberbullying techniques from different perspectives is discussed including data annotation, data preprocessing, and feature engineering. In addition, the importance of emojis in expressing emotions as well as their influence on sentiment classification and text comprehension leads us to discuss the role of incorporating emojis in the process of cyberbullying detection and their influence on the detection performance. Furthermore, the different domains for using self-supervised learning (SSL) as an annotation technique for cyberbullying detection are explored.

To Be Ethical and Responsible Digital Citizens or Not: A Linguistic Analysis of Cyberbullying on Social Media

As a worldwide epidemic in the digital age, cyberbullying is a pertinent but understudied concern—especially from the perspective of language. Elucidating the linguistic features of cyberbullying is critical both to preventing it and to cultivating ethical and responsible digital citizens. In this study, a mixed-method approach integrating lexical feature analysis, sentiment polarity analysis, and semantic network analysis was adopted to develop a deeper understanding of cyberbullying language. Five cyberbullying cases on Chinese social media were analyzed to uncover explicit and implicit linguistic features. Results indicated that cyberbullying comments had significantly different linguistic profiles than non-bullying comments and that explicit and implicit bullying were distinct. The content of cases further suggested that cyberbullying language varied in the use of words, types of cyberbullying, and sentiment polarity. These findings offer useful insight for designing automatic cyberbullying detection tools for Chinese social networking platforms. Implications also offer guidance for regulating cyberbullying and fostering ethical and responsible digital citizens.

The design, construction and evaluation of annotated Arabic cyberbullying corpus

Cyberbullying (CB) is classified as one of the severe misconducts on social media. Many CB detection systems have been developed for many natural languages to face this phenomenon. However, Arabic is one of the under-resourced languages suffering from the lack of quality datasets in many computational research areas. This paper discusses the design, construction, and evaluation of a multi-dialect, annotated Arabic Cyberbullying Corpus (ArCybC), a valuable resource for Arabic CB detection and motivation for future research directions in Arabic Natural Language Processing (NLP). The study describes the phases of ArCybC compilation. By way of illustration, it explores the corpus to discover strategies used in rendering Arabic CB tweets pulled from four Twitter groups, including gaming, sports, news, and celebrities. Based on thorough analysis, we discovered that these groups were the most susceptible to harassment and cyberbullying. The collected tweets were filtered based on a compiled harassment lexicon, which contains a list of multi-dialectical profane words in Arabic compiled from four categories: sexual, racial, physical appearance, and intelligence. To annotate ArCybC, we asked five annotators to classify 4,505 tweets into two classes manually: Offensive/non-Offensive and CB/non-CB. We conducted a rigorous comparison of different machine learning approaches applied on ArCybC to detect Arabic CB using two language models: bag-of-words (BoW) and word embedding. The experiments showed that Support Vector Machine (SVM) with word embedding achieved an accuracy rate of 86.3% and an F1-score rate of 85%. The main challenges encountered during the ArCybC construction were the scarcity of freely available Arabic CB texts and the deficiency of annotating the texts.

Curating Cyberbullying Datasets: a Human-AI Collaborative Approach

Cyberbullying is the use of digital communication tools and spaces to inflict physical, mental, or emotional distress. This serious form of aggression is frequently targeted at, but not limited to, vulnerable populations. A common problem when creating machine learning models to identify cyberbullying is the availability of accurately annotated, reliable, relevant, and diverse datasets. Datasets intended to train models for cyberbullying detection are typically annotated by human participants, which can introduce the following issues: (1) annotator bias, (2) incorrect annotation due to language and cultural barriers, and (3) the inherent subjectivity of the task can naturally create multiple valid labels for a given comment. The result can be a potentially inadequate dataset with one or more of these overlapping issues. We propose two machine learning approaches to identify and filter unambiguous comments in a cyberbullying dataset of roughly 19,000 comments collected from YouTube that was initially annotated using Amazon Mechanical Turk (AMT). Using consensus filtering methods, comments were classified as unambiguous when an agreement occurred between the AMT workers’ majority label and the unanimous algorithmic filtering label. Comments identified as unambiguous were extracted and used to curate new datasets. We then used an artificial neural network to test for performance on these datasets. Compared to the original dataset, the classifier exhibits a large improvement in performance on modified versions of the dataset and can yield insight into the type of data that is consistently classified as bullying or non-bullying. This annotation approach can be expanded from cyberbullying datasets onto any classification corpus that has a similar complexity in scope.

Multilingual Offensive Language Identification for Low-resource Languages

Offensive content is pervasive in social media and a reason for concern to companies and government organizations. Several studies have been recently published investigating methods to detect the various forms of such content (e.g., hate speech, cyberbullying, and cyberaggression). The clear majority of these studies deal with English partially because most annotated datasets available contain English data. In this article, we take advantage of available English datasets by applying cross-lingual contextual word embeddings and transfer learning to make predictions in low-resource languages. We project predictions on comparable data in Arabic, Bengali, Danish, Greek, Hindi, Spanish, and Turkish. We report results of 0.8415 F1 macro for Bengali in TRAC-2 shared task [23], 0.8532 F1 macro for Danish and 0.8701 F1 macro for Greek in OffensEval 2020 [58], 0.8568 F1 macro for Hindi in HASOC 2019 shared task [27], and 0.7513 F1 macro for Spanish in in SemEval-2019 Task 5 (HatEval) [7], showing that our approach compares favorably to the best systems submitted to recent shared tasks on these three languages. Additionally, we report competitive performance on Arabic and Turkish using the training and development sets of OffensEval 2020 shared task. The results for all languages confirm the robustness of cross-lingual contextual embeddings and transfer learning for this task.

Artificial Intelligence-Enabled Cyberbullying-Free Online Social Networks in Smart Cities

In recent years, smart city services have moved the existence of people from the physical to the virtual world (cyberspace), e.g., online banking, e-commerce, telemedicine, etc. Along with the benefits of smart cities, the problems of the physical world are also moved to the cyber world, like cyberbullying in online social networks (OSN). Automated cyberbullying detection techniques need to be designed to remove the potential tragedies in OSNs. The recent advent of artificial intelligence (AI) models like machine learning and deep learning (DL) models can be employed for the detection of cyberbullying in the OSN. With this motivation, this paper develops an AI-enabled cyberbullying-free OSN (AICBF-ONS) technique in smart cities. The proposed AICBF-ONS technique involves chaotic salp swarm optimization (CSSO)-based feature selection technique to derive a useful set of features from the OSN data. In addition, stacked autoencoder model is used as a classification model to allocate appropriate class labels of the OSN data. To improve the detection performance of the SAE model, a parameter tuning process take place using the mayfly optimization (MFO) algorithm. An extensive experimental analysis ensured the supremacy of the proposed AICBF-ONS technique.

“It is Typical of Teenagers”: When Teachers Morally Disengage from Cyberbullying

Teachers can contribute to preventing and solving cyberbullying situations. Therefore, it is relevant to investigate what may influence their involvement and actions concerning this phenomenon. A first study analyze teachers’ definitions of cyberbullying, how they would intervene and feel morally implicated with the phenomenon. A second study aimed to investigate the association between teachers’ being aware of cyberbullying and their perceived severity, moral disengagement with the phenomenon, perceived performance to solve such situations and their acquired knowledge about cyberbullying. Twenty semi-structured interviews were conducted in the first study with 25 to 65-year-old teachers. An online inventory was answered in study two by 541 middle and high school teachers (Mage = 50, SD = 7). A thematic analysis from the first study revealed that most teachers did not report repetition of behavior, power imbalance, intentionality to harm, and occurrence among peers as defining features of cyberbullying. Also, strategies they would use to intervene mainly focused on reporting the incident. Moreover, moral disengagement mechanisms were found in teachers’ discourse, which contribute to displacing responsibility for intervening and perceiving cyberbullying as less severe. In the second study, path analysis revealed that teachers’ awareness of cyberbullying among their students was positively associated with moral disengagement and acquired knowledge of the phenomenon. The mediating role of acquired knowledge of cyberbullying was significant between being aware of cyberbullying and teachers’ perceived severity of the situation, moral disengagement, and perceived performance to solve these situations. These findings highlight the relevance of developing cyberbullying training actions involving teachers.

Combating hate speech using an adaptive ensemble learning model with a case study on COVID-19

Social media platforms generate an enormous amount of data every day. Millions of users engage themselves with the posts circulated on these platforms. Despite the social regulations and protocols imposed by these platforms, it is difficult to restrict some objectionable posts carrying hateful content. Automatic hate speech detection on social media platforms is an essential task that has not been solved efficiently despite multiple attempts by various researchers. It is a challenging task that involves identifying hateful content from social media posts. These posts may reveal hate outrageously, or they may be subjective to the user or a community. Relying on manual inspection delays the process, and the hateful content may remain available online for a long time. The current state-of-the-art methods for tackling hate speech perform well when tested on the same dataset but fail miserably on cross-datasets. Therefore, we propose an ensemble learning-based adaptive model for automatic hate speech detection, improving the cross-dataset generalization. The proposed expert model for hate speech detection works towards overcoming the strong user-bias present in the available annotated datasets. We conduct our experiments under various experimental setups and demonstrate the proposed model's efficacy on the latest issues such as COVID-19 and US presidential elections. In particular, the loss in performance observed under cross-dataset evaluation is the least among all the models. Also, while restricting the maximum number of tweets per user, we incur no drop in performance.

FairPlay: Detecting and Deterring Online Customer Misbehavior

This study examines how firms can detect and manage customer misbehavior in online brand communities. We first develop a data science approach to detect customer misbehavior on social media and devise intervention strategies to deter it. Our design science approach achieves superior performance, improving detection by 7%–9% compared with traditional methods. We then implement two types of intervention policies based on injunctive (i.e., a punishment policy) and descriptive norms (i.e., a common identity policy) to restrain customer misbehavior. The results of field experiments indicate that punishment considerably reduces customer misbehavior in the short term, but this effect decays over time, whereas common identity has a smaller but more persistent effect on misbehavior reduction. In addition, punishing dysfunctional customers decreases their purchase frequency, whereas imposing a common identity increases it. Our results also show that combining the two policies effectively alleviates the detrimental effect of punishment, especially in the long run.

A Multichannel Deep Learning Framework for Cyberbullying Detection on Social Media

Online social networks (OSNs) play an integral role in facilitating social interaction; however, these social networks increase antisocial behavior, such as cyberbullying, hate speech, and trolling. Aggression or hate speech that takes place through short message service (SMS) or the Internet (e.g., in social media platforms) is known as cyberbullying. Therefore, automatic detection utilizing natural language processing (NLP) is a necessary first step that helps prevent cyberbullying. This research proposes an automatic cyberbullying method to detect aggressive behavior using a consolidated deep learning model. This technique utilizes multichannel deep learning based on three models, namely, the bidirectional gated recurrent unit (BiGRU), transformer block, and convolutional neural network (CNN), to classify Twitter comments into two categories: aggressive and not aggressive. Three well-known hate speech datasets were combined to evaluate the performance of the proposed method. The proposed method achieved promising results. The accuracy of the proposed method was approximately 88%.

Adolescent Deviance and Cyber-Deviance. A Systematic Literature Review

Deviance is a complex phenomenon that influences aspects both at the macro and micro levels, extensively studied by social scientists The main objective of this article was to conduct a systematic literature review for clustering the topics on adolescent deviance and online deviance. Grounded in Pickering's and Byrne's guidelines and PRISMA protocol, we identified the most recurrent themes, theories and predictors in the 61 most-cited articles related to the concept of deviance from the database of Web of Science, as well as in 488 abstracts of representative papers. The results emphasized four main clusters of topics, namely, predictors of deviance, online deviance, socio-constructivist theories, and research based theories of deviant behavior. The findings highlighted that researchers frequently use strain theory, social learning, self-control, and social control theories in their studies. Our systematic literature review revealed also the most encountered predictors of deviance, which we have classified into five main categories: family patterns, socio-demographic aspects, socialization, victimization, and school and individual factors. For online deviance, family patterns, socio-demographic aspects, victimization, school and individual factors, and Internet and computer use have been determined to be the main groups of predictors. The present systematic literature review makes an important contribution to the understanding of deviance by presenting an overview of the phenomenon.

Arabic Offensive and Hate Speech Detection Using a Cross-Corpora Multi-Task Learning Model

As social media platforms offer a medium for opinion expression, social phenomena such as hatred, offensive language, racism, and all forms of verbal violence have increased spectacularly. These behaviors do not affect specific countries, groups, or communities only, extending beyond these areas into people’s everyday lives. This study investigates offensive and hate speech on Arab social media to build an accurate offensive and hate speech detection system. More precisely, we develop a classification system for determining offensive and hate speech using a multi-task learning (MTL) model built on top of a pre-trained Arabic language model. We train the MTL model on the same task using cross-corpora representing a variation in the offensive and hate context to learn global and dataset-specific contextual representations. The developed MTL model showed a significant performance and outperformed existing models in the literature on three out of four datasets for Arabic offensive and hate speech detection tasks.

An Evaluation of Multilingual Offensive Language Identification Methods for the Languages of India

The pervasiveness of offensive content in social media has become an important reason for concern for online platforms. With the aim of improving online safety, a large number of studies applying computational models to identify such content have been published in the last few years, with promising results. The majority of these studies, however, deal with high-resource languages such as English due to the availability of datasets in these languages. Recent work has addressed offensive language identification from a low-resource perspective, exploring data augmentation strategies and trying to take advantage of existing multilingual pretrained models to cope with data scarcity in low-resource scenarios. In this work, we revisit the problem of low-resource offensive language identification by evaluating the performance of multilingual transformers in offensive language identification for languages spoken in India. We investigate languages from different families such as Indo-Aryan (e.g., Bengali, Hindi, and Urdu) and Dravidian (e.g., Tamil, Malayalam, and Kannada), creating important new technology for these languages. The results show that multilingual offensive language identification models perform better than monolingual models and that cross-lingual transformers show strong zero-shot and few-shot performance across languages.

Cyberbullying Detection in Social Networks Using Bi-GRU with Self-Attention Mechanism

With the propagation of cyberbullying in social networks as a trending subject, cyberbullying detection has become a social problem that researchers are concerned about. Developing intelligent models and systems helps detect cyberbullying automatically. This work focuses on text-based cyberbullying detection because it is the commonly used information carrier in social networks and is the widely used feature in this regard studies. Motivated by the documented success of neural networks, we propose a complete model combining the bidirectional gated recurrent unit (Bi-GRU) and the self-attention mechanism. In detail, we introduce the design of a GRU cell and Bi-GRU’s advantage for learning the underlying relationships between words from both directions. Besides, we present the design of the self-attention mechanism and the benefit of this joining for achieving a greater performance of cyberbullying classification tasks. The proposed model could address the limitation of the vanishing and exploding gradient problems. We avoid using oversampling or downsampling on experimental data which could result in the overestimation of evaluation. We conduct a comparative assessment on two commonly used datasets, and the results show that our proposed method outperformed baselines in all evaluation metrics.

Using Twitter to track immigration sentiment during early stages of the COVID-19 pandemic

Large-scale coordinated efforts have been dedicated to understanding the global health and economic implications of the COVID-19 pandemic. Yet, the rapid spread of discrimination and xenophobia against specific populations has largely been neglected. Understanding public attitudes toward migration is essential to counter discrimination against immigrants and promote social cohesion. Traditional data sources to monitor public opinion are often limited, notably due to slow collection and release activities. New forms of data, particularly from social media, can help overcome these limitations. While some bias exists, social media data are produced at an unprecedented temporal frequency, geographical granularity, are collected globally and accessible in real-time. Drawing on a data set of 30.39 million tweets and natural language processing, this article aims to measure shifts in public sentiment opinion about migration during early stages of the COVID-19 pandemic in Germany, Italy, Spain, the United Kingdom, and the United States. Results show an increase of migration-related Tweets along with COVID-19 cases during national lockdowns in all five countries. Yet, we found no evidence of a significant increase in anti-immigration sentiment, as rises in the volume of negative messages are offset by comparable increases in positive messages. Additionally, we presented evidence of growing social polarization concerning migration, showing high concentrations of strongly positive and strongly negative sentiments.

Current limitations in cyberbullying detection: On evaluation criteria, reproducibility, and data scarcity

The detection of online cyberbullying has seen an increase in societal importance, popularity in research, and available open data. Nevertheless, while computational power and affordability of resources continue to increase, the access restrictions on high-quality data limit the applicability of state-of-the-art techniques. Consequently, much of the recent research uses small, heterogeneous datasets, without a thorough evaluation of applicability. In this paper, we further illustrate these issues, as we (i) evaluate many publicly available resources for this task and demonstrate difficulties with data collection. These predominantly yield small datasets that fail to capture the required complex social dynamics and impede direct comparison of progress. We (ii) conduct an extensive set of experiments that indicate a general lack of cross-domain generalization of classifiers trained on these sources, and openly provide this framework to replicate and extend our evaluation criteria. Finally, we (iii) present an effective crowdsourcing method: simulating real-life bullying scenarios in a lab setting generates plausible data that can be effectively used to enrich real data. This largely circumvents the restrictions on data that can be collected, and increases classifier performance. We believe these contributions can aid in improving the empirical practices of future research in the field.

Multi-Class Imbalance in Text Classification: A Feature Engineering Approach to Detect Cyberbullying in Twitter

Twitter enables millions of active users to send and read concise messages on the internet every day. Yet some people use Twitter to propagate violent and threatening messages resulting in cyberbullying. Previous research has focused on whether cyberbullying behavior exists or not in a tweet (binary classification). In this research, we developed a model for detecting the severity of cyberbullying in a tweet. The developed model is a feature-based model that uses features from the content of a tweet, to develop a machine learning classifier for classifying the tweets as non-cyberbullied, and low, medium, or high-level cyberbullied tweets. In this study, we introduced pointwise semantic orientation as a new input feature along with utilizing predicted features (gender, age, and personality type) and Twitter API features. Results from experiments with our proposed framework in a multi-class setting are promising both with respect to Kappa (84%), classifier accuracy (93%), and F-measure (92%) metric. Overall, 40% of the classifiers increased performance in comparison with baseline approaches. Our analysis shows that features with the highest odd ratio: for detecting low-level severity include: age group between 19–22 years and users with <1 year of Twitter account activation; for medium-level severity: neuroticism, age group between 23–29 years, and being a Twitter user between one to two years; and for high-level severity: neuroticism and extraversion, and the number of times tweet has been favorited by other users. We believe that this research using a multi-class classification approach provides a step forward in identifying severity at different levels (low, medium, high) when the content of a tweet is classified as cyberbullied. Lastly, the current study only focused on the Twitter platform; other social network platforms can be investigated using the same approach to detect cyberbullying severity patterns.

A Descriptive Study of Bullying Victimization Among Arab American Adolescents in Southeast Michigan Middle and High Schools

PURPOSE

This cross-sectional study examined the frequency of different forms of bullying victimization (verbal, physical, and social), predictors of victimization, and whether bullying is reported to an adult.

DESIGN AND METHODS

The study utilized a community sample of 150 Arab American adolescents, age 12 to 16 years. The Adolescent Peer Relations Instrument-Victimization Scale was used to determine the participant's experiences of victimization in the past year. The adolescents indicated where bullying occurred, why, and whether they reported the incidence to an adult.

RESULTS

Approximately 30% of the study sample reported that victimization occurred occasionally (once a month or more frequent). Classrooms and hallways were the most common locations where bullying had occurred. Country-of-origin and obesity were the most frequent reasons for victimization. Predictors varied among the different forms of victimization; however, cyber-victimization [OR = 24.5; 95% CI 5-119.5)], perceived problematic attire [OR = 8.4; 95% CI 2.2-31.9)], female gender [OR = 5.2; 95% CI 1.2-22.7)], and being overweight [OR = 0.14; 95% CI 0.01-2.6)] all predicted overall victimization.

CONCLUSIONS

Our findings provide a foundation for future research focusing on Arab American adolescents, an underrepresented population, more research is needed to understand the scope of bullying victimization among Arab American adolescents.

PRACTICE IMPLICATIONS

This study will inform future intervention research and practice to consider victimization and related factors among Arab American adolescents. Culturally sensitive and multilevel interventions are imperative to decrease bullying victimization among Arab American adolescents and prevent negative effects on their health and families.

Cyberbullying severity detection: A machine learning approach

With widespread usage of online social networks and its popularity, social networking platforms have given us incalculable opportunities than ever before, and its benefits are undeniable. Despite benefits, people may be humiliated, insulted, bullied, and harassed by anonymous users, strangers, or peers. In this study, we have proposed a cyberbullying detection framework to generate features from Twitter content by leveraging a pointwise mutual information technique. Based on these features, we developed a supervised machine learning solution for cyberbullying detection and multi-class categorization of its severity in Twitter. In the study we applied Embedding, Sentiment, and Lexicon features along with PMI-semantic orientation. Extracted features were applied with Naïve Bayes, KNN, Decision Tree, Random Forest, and Support Vector Machine algorithms. Results from experiments with our proposed framework in a multi-class setting are promising both with respect to Kappa, classifier accuracy and f-measure metrics, as well as in a binary setting. These results indicate that our proposed framework provides a feasible solution to detect cyberbullying behavior and its severity in online social networks. Finally, we compared the results of proposed and baseline features with other machine learning algorithms. Findings of the comparison indicate the significance of the proposed features in cyberbullying detection.

Classification of Cyber-Aggression Cases Applying Machine Learning

The adoption of electronic social networks as an essential way of communication has become one of the most dangerous methods to hurt people’s feelings. The Internet and the proliferation of this kind of virtual community have caused severe negative consequences to the welfare of society, creating a social problem identified as cyber-aggression, or in some cases called cyber-bullying. This paper presents research to classify situations of cyber-aggression on social networks, specifically for Spanish-language users of Mexico. We applied Random Forest, Variable Importance Measures (VIMs), and OneR to support the classification of offensive comments in three particular cases of cyber-aggression: racism, violence based on sexual orientation, and violence against women. Experimental results with OneR improve the comment classification process of the three cyber-aggression cases, with more than 90% accuracy. The accurate classification of cyber-aggression comments can help to take measures to diminish this phenomenon.