Towards Detection of Cyberbullying in Hinglish Code Mixed Data

Covid-19 eruption and lockdown situation have increased the usages of online platforms which have impacted the users. Cyberbullying is one of the negative outcomes of using social media platforms which leads to mental and physical distress. This study proposes a machine learning-based approach for the detection of cyberbullying in Hinglish text. We use the Hinglish Code-Mixed Corpus, which consists of over 6,000 tweets, for our experiments. We use various machine learning algorithms, including Logistic regression (LR), Multinomial Naive Bayes (MNB), Support vector machine (SVM), Random Forest (RF), to train our models. We evaluate the performance of the models using standard evaluation metrics such as precision, recall, and F1-score. Our experiments show that the LR with Term Frequency-Inverse Document Frequency (TFIDF) outperforms the other models, achieving 92% accuracy. Our study demonstrates that machine learning models can be effective for cyberbullying detection in Hinglish text, and the proposed approach can help identify and prevent cyberbullying on social media platforms. © 2023 Bharati Vidyapeeth, New Delhi.

The Influence of Sharing Fake News, Self-Regulation, Cyber Bullying on Social Media Fatigue During COVID-19 Work Technology Conflict as Mediator Role

The purpose of this paper is to analyze the impact of Sharing Fake news, self-regulation, Cyber Bullying on social media fatigue during COVID-19 work technology conflict as mediator role. The current study uses quantitative with cross sectional design to examine the effect of Sharing Fake news, self-regulation, Cyber Bullying on social media fatigue during COVID-19 and suing work technology conflict as mediator role. The respondents were situation from different top sites, such as twitter, Facebook and Instagram sample of 132, and population of this study is 200 users were selected for this study, sample size is calculated through ROA soft. The dissemination of unverified information has been showcased as a significant challenge during the COVID-19 pandemic. The role of social media in this process is exemplified by its increased use during COVID-19, as, for example, a recent report shows that the use of Facebook hit record levels during the pandemic. This study revealed that potential mechanisms for counteracting fake news creating Facebook pages of real news and using this advertising to disseminate accurate information this paper will enhanced the understanding the effects of SMA, cyberbullying and self-regulation on mental health of individual through the use of social cognitive perspectives. To enhance efficacy in the role of social media in this process to reduce the gap between theory and practice, social marketers should include messages that empower recipients. Campaigns should show recommended behaviors and highlight their usefulness and effectiveness. This paper has been methodological as well as theoretical limitations, first using CLT might be regarded limiting even through it has been adopted in past studies that are based on social media. The primary concern is the cognitive load theory is still essential an instructional science theory even through it has been adopted, use widely in HCI & also shown to explain not only learning but also acquiring knowledge from new theories, articles might be more useful for conceptualizing fake news sharing. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Investigating User Information and Social Media Features in Cyberbullying Detection

As society grows increasingly more online with each passing year, the problem of cyberbullying becomes more and more prominent, with such incidents having the capacity to negatively impact mental health in a major way, especially among children and teenagers. The proposed approach builds on our previous work that established multi-modal detection of cyberbullying on Twitter, and restructures the multi-modal approach by incorporating social media features such as time-related features and social network information. As a result, the new models reach a classification accuracy between 94.4% and 94.6%, from the previous accuracy of 93%. The proposed new approach affirms the use of context-based data in addition to more directly-related features when analyzing cyberbullying and other interactions with promising improvements. We believe that this work contributes significantly to the study of cyberbullying detection, which is an imminent problem with growing importance in the post-COVID society. © 2022 IEEE.

LuMee: A Centralized Smart Protector for School Children who are Using Online Education

This study was motivated by the challenges experienced by parents and guardians in ensuring the safety of children in cyberspace. In the last two or three years, online education has become very popular all over the world due to the Covid 19 pandemic. Therefore parents, guardians and teachers must ensure the safety of children in cyber space. Children are more likely to go astray and there are plenty of online programs waiting to get them on wrong track and also children who are engaging in the online education can be distracted at any moment. Therefore, parents should keep a close check on their children's online activity. Apart from that due to the unawareness of children, they tempt to share their sensitive information, chance of being a victim of phishing attacks from outsiders. These problems can be overcome through the proposed web-based system. We use feature extraction, web tracking and analysis mechanisms, image processing and name entity recognition to implement this web-based system. © 2022 IEEE.

Exploring Crowdsourced Content Moderation Through Lens of Reddit during COVID-19

In 2020, when COVID-19 struck, social media gained even more influence in people's lives due to increased online activity. This event led to a surge of false information and cyberbullying, making content moderation harder than ever. Given this challenge, exploring opportunities to explore content moderation solutions to reduce hate speech and fake news on social media is vital. In this paper, we examine if existing content moderation systems are enough during global pandemics and, if not, where gaps may lie. Due to its intriguing Decentralized Content Management System (DCMS), we chose Reddit as the key social networking platform for our hypothesis testing. We used 1.8 million Reddit posts from COVID-19-related subreddits from January 2020 to April 2021. Our findings reveal several significant trends regarding the effect of a worldwide event on content moderation methods designed to lessen the prevalence of hazardous content and fake news. In light of these considerations, we provide the results of comprehensive research conducted with particular attention paid to the user-generated material and the DCMS of Reddit. © 2022 ACM.

Towards a Method to Create Microlearning Capsules: Bullying and Cyberbullying Domain

The development of new information and communication technologies (ICTs) has become widespread due to the technology development and the COVID-19 pandemic;consequently, internet use has increased and impacted different social areas, like communication, education, and the way people relate. Therefore, increasing the cases of bullying and cyberbullying in children and teenagers. In addition, instructional tools, such as microlearning, have emerged in education, allowing for compressing and massifying educational content from different areas of knowledge. This tool aims to reduce learning time and can be deployed on any platform without using traditional means of learning. This paper presents an overview of a method for creating learning capsules in the domain of bullying and cyberbullying named LeCCMe. This method is based on the ADDIE instructional model and incorporates a diffusion phase to reach the largest possible population. Finally, a case study is presented, in which a learning capsule has been created to prevent bullying and cyberbullying. © 2022 IEEE.

Too much information: exploring technology-mediated abuse in higher education online learning and teaching spaces resulting from COVID-19 and emergency remote education.

During COVID-19, universities across the globe experienced a rapid requirement to move to online learning and teaching provision. This rapid move has been explored as emergency remote education (ERE). This paper reviews and presents some emerging literature regarding ERE, demonstrating how this created an environment where technology-mediated abuse could arise within the university context. Intentional and unintentional forms of technology-mediated abuse, within a global context, are considered with account of how intersectional characteristics can impact. The paper concludes with a set of provocations explored within an example framework. The provocations are given to situate ways of thinking which are facilitative of safer and more respectful use of technological spaces. Both the provocations and example framework aim to be useful critical tools for program and module teams to adapt in higher education institutions within the online sphere. The phenomenon of ERE is an opportunity to consider what can be learned with regard to management of technology-mediated abuse. However, a focus on ERE presents limitations in the paper because of the smaller number of academic sources at this time, due to recency of the COVID-19 pandemic.

Forensic Acquisition and Analysis of Webpage

Due to the onset of the Covid-19 pandemic, people are compelled to maintain social distance in all spheres of life, forcing people to adopt virtual mode of activity. Usage of social media and other internet activity has shot up in this period, and consequently, cybercrimes have also increased. If cybercrimes are reported, computer forensics analysts will examine the concerned website, online forum, or social media to find meticulous details about the cybercrime. But webpage content seen on the day may not be available on the next day. The contents of the webpage, which is the subject of crime, will be deleted or withdrawn, or deactivated to destroy evidence to escape from legal proceedings. The victims usually produce a screenshot of the webpage or image or video as a piece of evidence. But there is a distinct possibility of manipulating the offensive materials and it may not be considered a valid piece of evidence before the court of law. Such a scenario requires a forensic technique that should acquire the content of the webpage before it is removed from web site to maintain the authenticity of captured data. So, we are proposing an automated system for the forensic acquisition of a website that will effectively capture all content from the live website and make it useful for forensic investigation and may be produced before the court as valid evidence of cybercrime. © 2022 IEEE.

Hate Speech Detection in Thai Social Media with Ordinal-Imbalanced Text Classification

Cyberbullying has become a serious problem in Thai social media. For example, some Thai people posted hate speeches on Myanmar workers in Thailand during the COVID-19 pandemic, which might elevate hate crime. It is imperative and urgent to detect cyberbullying on Thai social media. The task is a text classification problem. Moreover, hate speeches contain the order of severity levels, but many pieces of work did not consider this point in the model. Therefore, we developed a Thai hate-speech classification method with various loss functions to detect such hate speeches accurately. We evaluated them on a corpus of ordinal-imbalanced Thai text. The evaluated outcomes indicated that the best-in terms of $F$1 -score-model was the model with a loss function of a hybrid between an Ordinal regression loss function and Pearson correlation coefficients (common in similarity function). It yielded an average F1-score of 78.38 %-0.88 % significantly higher than the score achieved by a conventional loss function-and an average mean squared error of 0.2478-5.49 % relative improvement. Thus, the proposed hybrid loss function improved the efficiency of the model. © 2022 IEEE.

Problematic mobile phone use and cyberbullying perpetration in adolescents

The phenomenon of problematic mobile phone use has become increasingly common among adolescents during the lockdowns mandated by the COVID-19 pandemic. However, research is still scarce on the impact of such use on delinquent cyberspace conduct (i.e. cyberbullying). This study applies the theoretical framework of general strain theory to examine how problematic mobile phone use affects the perpetration of cyberbullying. The results of this empirical examination of longitudinal survey data obtained from 2,161 adolescents in South Korea reveal that problematic mobile phone use is positively associated with engagement in cyberbullying. It is a type of strain that induces negative emotional states and results in the perpetration of cyberbullying. Furthermore, this study investigates the moderating roles of both traditional bullying experiences (i.e. traditional bullying and victimisation) in the association between problematic mobile phone use and the perpetration of cyberbullying. We found traditional bullying perpetration positively moderates the effects of problematic mobile phone use on cyberbullying. On the other hand, we found the moderating effect of traditional bullying victimisation of adolescents was insignificant. © 2022 Informa UK Limited, trading as Taylor & Francis Group.

Cyberbullying Detection with Side Information: A Real-World Application of COVID-19 News Comment in Chinese Language

Cyberbullying is an aggressive and intentional behavior committed by groups or individuals, and its main manifestation is to make offensive or hurtful comments on social media. The existing researches on cyberbullying detection underuse natural language processing technology, and is only limited to extracting the features of comment content. Meanwhile, the existing datasets for cyberbullying detection are non-standard, unbalanced, and the data content of datasets is relatively outdated. In this paper, we propose a novel Hybrid deep Model based on Multi-feature Fusion (HMMF), which can model the content of news comments and the side information related to net users and comments simultaneously, to improve the performance of cyberbullying detection. In addition, we present the JRTT: a new, publicly available benchmark dataset for cyberbullying detection. All the data are collected from social media platforms which contains Chinese comments on COVID-19 news. To evaluate the effectiveness of HMMF, we conduct extensive experiments on JRTT dataset with five existing pre-trained language models. Experimental results and analyses show that HMMF achieves state-of-the-art performances on cyberbullying detection. To facilitate research in this direction, we release the dataset and the project code at https://github.com/xingjian215/HMMF. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Cyberbullying detection system focusing on the isiXhosa language

Due to the Covid19 pandemic, and the restrictions placed in social interactions, there has been an upsurge in the use of social networks, such as Facebook, WhatsApp, Instagram, Telegram, Twitter, and others. As more people turn to the social networks for social interaction, there has been increased occurrences of cyberbullying. Cyberbullying is a type of bullying that occurs through online technology, whereby harmful texts and pictures are shared through social networks. This research project aimed to develop a system that can detect cyberbullying on social networks such as Twitter focusing on the IsiXhosa language. Machine learning algorithms were applied to Twitter feeds in order to detect cyberbullying. The project will help law enforcement to apprehend and prosecute cyberbullies that make threats using isiXhosa. The methodology used incorporated machine learning algorithms to fully implement the cyberbullying detection system. It starts with collecting the data from Twitter using Python, cleaning the data followed by testing the data. The results show that the implementation successfully collected the desired data from Twitter and the data was then pre-processed and prepared to be tested using the different algorithms mentioned in the paper. © 2022 IEEE.

Analysing Cyberbullying Using Natural Language Processing by Understanding Jargon in Social Media

Cyberbullying is of extreme prevalence today. Online-hate comments, toxicity, and cyberbullying amongst vulnerable groups is only growing over increased access to social platforms, especially post COVID-19. It is paramount to detect and ensure safety across social platforms so that any violence or hate-crime is automatically detected and strict action is taken against it. In our work, we explore binary classification by using a combination of datasets from various social media platforms that cover a wide range of cyberbullying such as sexism, racism, abusive, and hate-speech. We experiment through multiple models such as Bi-LSTM, GloVe, state-of-the-art models like BERT, and apply a unique preprocessing technique by introducing a slang-abusive corpus, achieving a higher precision in comparison to models without slang preprocessing. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Marketing Process in Information Security Context: Comparison Between Czech Republic and Belgium

Due to increased numbers of cyber-attacks, cyberbullying and data breaches incidents influence the Digital Marketing Process that refers to the processes involved in delivering products and services to the customer (process of paying for influencing practically in the same way as the value users received). The unreliability and extreme information vulnerability of Digital marketing processes, as well as implemented collaboration platforms that complete with different document sharing, online meeting modules, etc., which are part of a company's operational marketing infrastructure processes and marketing mix tools, afterwards leads to a loss of the company's reputation and, as a result, to financial losses. The attitude to the issue of information security and personal data protection is different from country to country and from business to business thus represent different ways. This paper proves that there is a relationship between the use of information security strategy as part of the digital marketing process and the company's positive reputation among consumers in countries such Belgium and the Czech Republic, as well as the willingness of businesses in these two countries to invest resources in information security to reinforce the marketing process. The study provides evidence that there is a difference in the perceptions of information security among businesses and consumers in Belgium and the Czech Republic, and differences in the perception of information security threats and data privacy before the Coronavirus and current attitude of future strategies in relation of information security as good reputation guaranty of the reliable marketing mix process. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Am I Being Bullied on Social Media? An Ensemble Approach to Categorize Cyberbullying

People can easily reveal their aggressive remarks on social media platforms using the anonymity it provides. During the COVID-19 pandemic, the usage of social media has been increased several times according to surveys and people are vulnerable to cyber attacks now more than ever. Prevention of cyberbullying needs careful monitoring and identification. Most of the existing works on cyberbullying detection employed traditional machine learning classifiers with handcrafted fea-tures, and deep learning-based models have made their way in this domain very recently. Categorizing cyberbullying based on traits is a complex task and needs contextual consideration. In this work, we have proposed a new approach to detect cyberbullying on social media platforms using a neural ensemble method of transformer-based architectures with attention mechanism. Our proposed architecture is trained on one balanced and one imbalanced dataset and outperforms the given ML and DNN baselines by a significant margin in both cases. We achieved an average F1-score of 95.59% for five classes and 90.65% for six classes on the Fine-Grained Cyberbullying Dataset (FGCD), and 87.28% on Twitter parsed dataset. Our in-depth results provide great insights into the effectiveness of transformer-based models in cyberbullying detection and paves the way for future researches to combat this serious online issue. We have released our models and code.1 © 2021 IEEE.

Denigrate Comment Detection in Low-Resource Hindi Language Using Attention-Based Residual Networks

Cyberspace has been recognized as a conducive environment for use of various hostile, direct, and indirect behavioural tactics to target individuals or groups. Denigration is one of the most frequently used cyberbullying ploys to actively damage, humiliate, and disparage the online reputation of target by sending, posting, or publishing cruel rumours, gossip, and untrue statements. Previous pertinent studies report detecting profane, vulgar, and offensive words primarily in the English language. This research puts forward a model to detect online denigration bullying in low-resource Hindi language using attention residual networks. The proposed model Hindi Denigrate Comment-Attention Residual Network (HDC-ARN) intends to uncover defamatory posts (denigrate comments) written in Hindi language which stake and vilify a person or an entity in public. Data with 942 denigrate comments and 1499 non-denigrate comments is scraped using certain hashtags from two recent trending events in India: Tablighi Jamaat spiked Covid-19 (April 2020, Event 1) and Sushant Singh Rajput Death (June 2020: Event 2). Only text-based features, that is, the actual content of the post, are considered. The pre-Trained word embedding for Hindi language from fastText is used. The model has three ResNet blocks with an attention layer that generates a post vector for a single input, which is passed through a sigmoid activation function to get the final output as either denigrate (positive class) or non-denigrate (negative class). An F-1 score of 0.642 is achieved on the dataset. © 2021 Association for Computing Machinery.

CYBERBULLYING in ONLINE CLASSES: The CASE of COVID-19

The digital economy has evolved due to the Covid-19 pandemic. Like businesses, the educational sector is also affected by this change. Students are using online educational platforms to continue their studies during the pandemic. The digital world provides resilience but it has also caused trust and privacy issues for students. Cyberbullying has significantly increased in online classrooms during the pandemic and it is essential to understand the increased risks of cyberbullying in online classrooms. It is vital to understand its effect on students. This study explores the impacts ofcyber violence and cyber oppression that students face in online classrooms. An online survey is conducted to gain knowledge of (i) cyberbullying in online classrooms, (ii) its negative impacts on students, and (iii) the lack of awareness, policies and management regarding cyber laws in educational institutions under the pandemic. Based on this study's finding, students'online safety is at risk. Proper anti-cyberbullying policies are almost non-existent, and there is a lack of awareness about existinganti-cyberbullying laws. This survey also highlights the lack of research regarding successful practices, strategies, anti-cyberbullying policies, and legal reactions for cyberbullying victims. Therefore, a cyberbullying prevention framework is proposed in which reactive and preventive measures and educational institutions' responsibilities are suggested. The proposed framework will help to formulate cyberbullying anticipation programs for online educational platforms and it will also help in minimising students' trust and privacy concerns due to digitalised educational systems. © 2021 Institution of Engineering and Technology. All rights reserved.

Going Viral: Researching Safely on Social Media.

Safety issues for researchers conducting and disseminating research on social media have been inadequately addressed in institutional policies and practice globally, despite posing significant challenges to research staff and student well-being. In the context of the COVID-19 pandemic and given the myriad of advantages that web-based platforms offer researchers over traditional recruitment, data collection, and research dissemination methods, developing a comprehensive understanding of and guidance on the safe and effective conduct of research in web-based spaces has never been more pertinent. In this paper, we share our experience of using social media to recruit participants for a study on abortion stigma in Australia, which brought into focus the personal, professional, and institutional risks associated with conducting web-based research that goes viral. The lead researcher (KV), a postgraduate student, experienced a barrage of harassment on and beyond social media. The supportive yet uncoordinated institutional response highlighted gaps in practice, guidance, and policy relating to social media research ethics, researcher safety and well-being, planning for and managing web-based and offline risk, and coordinated organizational responses to adverse events. We call for and provide suggestions to inform the development of training, guidelines, and policies that address practical and ethical aspects of using social media for research, mental and physical health and safety risks and management, and the development of coordinated and evidence-based institutional- and individual-level responses to cyberbullying and harassment. Furthermore, we argue the case for the urgent development of this comprehensive guidance around researcher safety on the web, which would help to ensure that universities have the capacity to maximize the potential of social media for research while better supporting the well-being of their staff and students.

Receiving threatening or obscene messages from a partner and mental health, self-harm and suicidality: results from the Adult Psychiatric Morbidity Survey.

PURPOSE: Threatening or obscene messaging is repeated, unwanted texts, emails, letters or cards experienced by the recipient as threatening or obscene, and causing fear, alarm or distress. It is rarely examined as an aspect of intimate partner violence. We describe the prevalence of exposure to threatening/obscene messaging from a current or ex-partner; characteristics of victims; and associations with other forms of violence and abuse, mental disorder, self-harm, and suicidality. METHODS: Cross-sectional probability-sample survey of the general population in England aged 16 + . Multivariable regression modelling tested associations between receipt of threatening/obscene messaging and current common mental disorder, past-year self-harm and suicidality. RESULTS: Threatening/obscene messages were received from a current/ex-partner by 6.6% (95%CI: 5.9-7.3) of adults who had been in a relationship; 1.7% received these in the past year. Victims were more likely to be female, under 35, single or divorced, socioeconomically disadvantaged, and to have experienced other forms of sexual and partner violence and abuse. Those who received threatening/obscene messages in the past year were more likely to experience common mental disorder (adjusted odds ratio 1.89; 1.01-3.55), self-harm (2.31; 1.00-5.33), and suicidal thoughts (2.00; 1.06-3.78). CONCLUSION: Threatening/obscene messaging commonly occurs in the context of intimate partner violence. While often occurring alongside sexual and physical violence, messaging has an additional association with mental disorder and suicidality. Routine enquiry in service settings concerning safety, including those working with people who have escaped domestic violence, should ask about ongoing contact from previous as well as current partners. This should include asking about messaging, as well as other forms of potentially technology-enabled abuse which may become increasingly common.