Cyberbullying detection on twitter using Big Five and Dark Triad features

Threatening language detection from Urdu data with deep sequential model

The Urdu language is spoken and written on different social media platforms like Twitter, WhatsApp, Facebook, and YouTube. However, due to the lack of Urdu Language Processing (ULP) libraries, it is quite challenging to identify threats from textual and sequential data on the social media provided in Urdu. Therefore, it is required to preprocess the Urdu data as efficiently as English by creating different stemming and data cleaning libraries for Urdu data. Different lexical and machine learning-based techniques are introduced in the literature, but all of these are limited to the unavailability of online Urdu vocabulary. This research has introduced Urdu language vocabulary, including a stop words list and a stemming dictionary to preprocess Urdu data as efficiently as English. This reduced the input size of the Urdu language sentences and removed redundant and noisy information. Finally, a deep sequential model based on Long Short-Term Memory (LSTM) units is trained on the efficiently preprocessed, evaluated, and tested. Our proposed methodology resulted in good prediction performance, i.e., an accuracy of 82%, which is greater than the existing methods.

Agreeableness and adolescents' cyberbullying perpetration: A longitudinal moderated mediation model of moral disengagement and empathy

OBJECTIVE

The current study explored whether agreeableness predicted cyberbullying perpetration across 3 years and extended previous studies by exploring the mediating effect of moral disengagement and the moderating effects of empathy and gender.

METHOD

The participants included 2407 adolescents from 7 middle schools in China. They were recruited to complete the Big Five Personality Inventory, Bullying Scale and Empathy Scale at Time 1, Moral Disengagement Scale at Time 1 and Time 2, and Cyberbullying Perpetration Scale at Time 1, Time 2, and Time 3.

RESULTS

Agreeableness at Time 1 predicted cyberbullying perpetration at Time 3 and moral disengagement at Time 2 mediated this relationship. The relationship between moral disengagement at Time 2 and cyberbullying perpetration at Time 3 was stronger for low cognitive empathy adolescents than high cognitive empathy adolescents at Time 1. The relationship between agreeableness at Time 1 and cyberbullying perpetration adolescents at Time 3 was stronger for low affective empathy than high affective empathy adolescents at Time 1. The link between moral disengagement at Time 2 and cyberbullying perpetration at Time 3 was weaker for females than males.

CONCLUSIONS

Low agreeableness adolescents are more likely to use moral disengagement, which in turn leads to more cyberbullying perpetration.

Civic Engagement and Personality: Associations with the Big Five and the Dark Triad

Several studies have analyzed the relationship between general personality traits and attitudes and behaviors, indicating that a person is more committed to the community. After raising the question of whether malevolent traits might also be related, the aim was to analyze the relationship between civic engagement and personality, delving into the contribution of the Dark Triad (narcissism, Machiavellianism, and psychopathy) and controlling for the association with the Big Five. The Civic Engagement Questionnaire, the Short Dark Triad, and the Big Five Inventory-10 were administered to 1175 Spanish students (convenience sampling). After performing statistical analyses using SPSS statistical software, it was obtained that the three Dark Triad traits explained 11% of the total explained variance of civic engagement, while 19% was reached when the Big Five were included. Narcissism and openness were the factors most strongly associated with engagement. The positive relationship between narcissism and general personality traits could explain why narcissistic people have more favorable attitudes. Furthermore, people with narcissistic traits may display these attitudes for their own benefit. This study provides further evidence of how the narcissistic personality trait differs from the other two malevolent traits. Given that these traits are also associated with maladaptive behaviors, knowing all their characteristics could facilitate the design of prevention programs aimed at reducing such maladaptive behaviors.

Dark tetrad of personality, cyberbullying, and cybertrolling among young adults

Communication applications and social media sites serve as a platform for users to distribute information and connect to other users, potentially allowing perpetrators to perform antisocial behaviors. The current study examined the relationship between Dark Tetrad of personality (i.e., Machiavellianism, narcissism, psychopathy, sadism) and antisocial cyber-behaviors (i.e., cyberbullying, cybertrolling) by surveying young Malaysians (n = 323) aged from 18 to 26. Partial least squares structural equation modelling (PLS-SEM) revealed that Machiavellianism was not related to cyberbullying and cybertrolling, while narcissism was positively related to cyberbullying but not related to cybertrolling. Meanwhile, psychopathy and sadism were positively related to cyberbullying and cybertrolling. The results of this study contribute to the cyber-behaviors literature, knowledge about the antisocial cyber-behaviors in Malaysia, supports sadism as a dark personality and the study acts as a reference to minimize these behaviors.

Identification of cyber harassment and intention of target users on social media platforms

Due to Coronavirus diseases in 2020, all the countries departed into lockdown to combat the spread of the pandemic situation. Schools and institutions remain closed and students' screen time surged. The classes for the students are moved to the digital platform which leads to an increase in social media usage. Many children had become sufferers of cyber harassment which includes threatening comments on young students, sexual torture through a digital platform, people insulting one another, and the use of fake accounts to harass others. The rising effort on automated cyber harassment detection utilizes many AI-related components Natural language processing techniques and machine learning approaches. Though machine learning models using different algorithms fail to converge with higher accuracy, it is much more important to use significant natural language processes and efficient classifiers to detect cyberbullying comments on social media. In this proposed work, the lexical meaning of the text is analysed by the conventional scheme and the word order of the text is performed by the Fast Text model to improve the computational efficacy of the model. The intention of the text is analysed by various feature extraction methods. The score for intention detection is calculated using the frequency of words with a bully-victim participation score. Finally, the proposed model's performance is measured by different evaluation metrics which illustrate that the accuracy of the model is higher than many other existing classification methods. The error rate is lesser for the detection model.

Personality of nonprofit organizations’ Instagram accounts and its relationship with their photos’ characteristics at content and pixel levels

Nonprofit organizations (NPO) can utilize social networking sites (SNSs) for their activities. Like individual users, they can create SNS accounts, upload posts to show what they are doing, and communicate with other users. Thus, their accounts can be investigated from the same perspective of personality which has been one of the key lenses through which SNS posts of individual users was investigated. In the line of literature that analyzed the personality of non-human objects such as products, stores, brands, and websites, the present research analyzed the personality of NPOs’ Instagram accounts using an online AI service. Also, it investigated how their personality traits were related to the characteristics of the uploaded photos at content and pixel levels. The results of analysis of 223,446 photos on 177 Instagram accounts suggested that the personality of NPOs’ Instagram accounts can be summarized as being high in openness and agreeableness but low in extraversion and neuroticism. And it was found that openness and agreeableness were the personality traits that associated the most with the photo features. Also, the personality traits of NPOs’ Instagram accounts, except neuroticism, were predicted from the photo features with an acceptable level of accuracy. Implications of this research and suggestions for further research were presented.

A survey of uncover misleading and cyberbullying on social media for public health

Misleading health information is a critical phenomenon in our modern life due to advance in technology. In fact, social media facilitated the dissemination of information, and as a result, misinformation spread rapidly, cheaply, and successfully. Fake health information can have a significant effect on human behavior and attitudes. This survey presents the current works developed for misleading information detection (MLID) in health fields based on machine learning and deep learning techniques and introduces a detailed discussion of the main phases of the generic adopted approach for MLID. In addition, we highlight the benchmarking datasets and the most used metrics to evaluate the performance of MLID algorithms are discussed and finally, a deep investigation of the limitations and drawbacks of the current progressing technologies in various research directions is provided to help the researchers to use the most proper methods in this emerging task of MLID.

Relationship between Honesty-Humility and Cyberbullying Perpetration: A Moderated Mediation Model

Previous studies have verified the roles of big five personalities in cyberbullying perpetration (CP). The Big Five model has been revised to include an additional dimension, called Honesty-Humility (HH). It is not clear whether HH would be associated with CP. Thus, the effect of HH on CP was examined. To further explore this influencing mechanism, materialism was examined as a mediator, and parental psychological control (PPC) was examined as a moderator in the relationship between HH and CP. A total of 1,004 Chinese adolescents (M = 12.95, SD =1.12) participated in this study using a cross-sectional design and multiple questionnaires, namely, the Honesty-Humility subscale of the 24-item Brief HEXACO Inventory, the Revised Cyber Bullying Inventory, the Material Values Scale for Children, and the Parental Control Questionnaire. Correlation analyses indicated that CP, materialism, and PPC were significantly and positively correlated with each other, and were significantly and negatively associated with HH. The mediation model revealed that materialism played a mediating role in the relationship between HH and CP. The moderated mediation model revealed that stronger PPC strengthened the direct associations of HH with materialism and CP, and further strengthened the indirect relationship between HH and CP. Specifically, Chinese adolescents with lower levels of HH were more likely to build material values and further engaged in cyberbullying perpetration, when they perceived stronger PPC.

How many cyberbullying(s)? A non-unitary perspective for offensive online behaviours

Research has usually considered cyberbullying as a unitary phenomenon. Thus, it has been neglected to explore whether the specific online aggressive behaviours relate differentially to demographic features of the perpetrators of online aggressive actions, their personality characteristics, or to the ways in which they interact with the Internet. To bridge this gap, a study was conducted through a questionnaire administered online to 1228 Italian high-school students (Female: 61.1%; 14–15 yo: 48.%; 16–17 yo: 29.1%; 18–20 yo: 20.4%, 21–25 yo: 1.6%; Northern Italy: 4.1%; Central Italy: 59.2%; Southern Italy: 36.4%). The questionnaire, in addition to items about the use of social media, mechanisms of Moral Disengagement and personality characteristics of the participants in the study, also included a scale for the measurement of cyberbullying through the reference to six aggressive behaviours. The results indicate that cyberbullying can be considered as a non-unitary phenomenon in which the different aggressive behaviours can be related to different individual characteristics such as gender, personality traits and the different ways of interacting with social media. Moreover, the existence of two components of cyberbullying has been highlighted, one related to virtual offensive actions directly aimed at a victim, the other to indirect actions, more likely conducted involving bystanders. These findings open important perspectives for understanding, preventing, and mitigating cyberbullying among adolescents.

The Dark Tetrad, cybervictimization, and cyberbullying: The role of moral disengagement

Cyberbullying is a form of antisocial online behaviors. Perpetration of intentional and repeated harm inflicted through electronic devices is associated with dark personality traits and may be caused by morally impaired reasoning. In the current study, we investigated the associations between the Dark Tetrad (narcissism, Machiavellianism, psychopathy, sadism), cybervictimization, and cyberbullying. We also examined the intervening role of moral disengagement in the relationship between the Dark Tetrad and cyberbullying. Two hundred fifty-one adults (72.6% women) participated in an on-line study. Correlational analysis indicated that all dark personality traits were associated with higher cyberbullying and cybervictimization (except narcissism as a predictor of cybervictimization). Moral disengagement was positively related to Machiavellianism, sadism and cybervictimization. Controlled for covariance between the Dark Tetrad traits and cybervictimization, sadism and cybervictimization appeared to be associated with cyberbullying. Moreover, moral disengagement did not account for the associations between the Dark Tetrad and cyberbullying.

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.

Artificial Intelligence to Address Cyberbullying, Harassment and Abuse: New Directions in the Midst of Complexity

This brief article serves as an introductory piece for the special issue “The Use of Artificial Intelligence (AI) to Address Online Bullying and Abuse.” It provides an overview of the state of the art with respect to the use of AI in addressing various types of online abuse and cyberbullying; current challenges for the field; and it emphasises the need for greater interdisciplinary collaboration on this topic. The article also summarises key contributions of the articles selected for the special issue.

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%.

Preference for Number of Friends in Online Social Networks

Preferences or dislikes for specific numbers are ubiquitous in human society. In traditional Chinese culture, people show special preference for some numbers, such as 6, 8, 10, 100, 200, etc. By analyzing the data of 6.8 million users of Sina Weibo, one of the largest online social media platforms in China, we discover that users exhibit a distinct preference for the number 200, i.e., a significant fraction of users prefer to follow 200 friends. This number, which is very close to the Dunbar number that predicts the cognitive limit on the number of stable social relationships, motivates us to investigate how the preference for numbers in traditional Chinese culture is reflected on social media. We systematically portray users who prefer 200 friends and analyze their several important social features, including activity, popularity, attention tendency, regional distribution, economic level, and education level. We find that the activity and popularity of users with the preference for the number 200 are relatively lower than others. They are more inclined to follow popular users, and their social portraits change relatively slowly. Besides, users who have a stronger preference for the number 200 are more likely to be located in regions with underdeveloped economies and education. That indicates users with the preference for the number 200 are likely to be vulnerable groups in society and are easily affected by opinion leaders.

Cyberbullying and cyber-mobbing in developing countries

Purpose

This paper aims to consider modern internet phenomena such as cyberbullying and cybermobbing. The emphasis in the paper is placed on the problematic issues of the legal practice of combating cyberbullying and cyber-mobbing in developing countries as these phenomena are still insufficiently studied. The subject of this paper is modern internet phenomena such as cyberbullying and cyber-mobbing. The emphasis in the paper is placed on the problematic issues of the legal practice of combating cyberbullying and cyber-mobbing in developing countries as these phenomena are still insufficiently studied.

Design/methodology/approach

The legislation of developing countries is compared with doctrinal and practical developments in the fight against the studied problem in developed countries of the West, as well as countries of the former USSR. Moreover, experiment was conducted to determine the effectiveness of methods to combat cyberbullying using social networks. Thus, 40 random accounts of people (presumably from 18 to 30 years old) were analyzed.

Findings

This paper indicates the concepts of cyber-mobbing and cyberbullying, as well as their varieties that exist in the modern world. This study examines statistical data, programs and measures of different states in the fight against cyberbullying and cyber-mobbing. Results of experiments showed that Instagram users are aware of the availability of built-in extensions of the social network to protect against cyberbullying and use them relatively frequently. With that, female segment of Instagram users is more concerned about the content of the comments under their photos than the male one.

Originality/value

Measures have been developed to prevent and counteract cyberbullying and cyber-mobbing, the introduction of which into the policies of states might help in the fight against these social phenomena.

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.

Personality of Public Health Organizations' Instagram Accounts and According Differences in Photos at Content and Pixel Levels

Organizations maintain social media accounts and upload posts to show their activities and communicate with the public, as individual users do. Thus, organizations’ social media accounts can be examined from the same perspective of that of individual users’ accounts, with personality being one of the perspectives. In line with previous studies that analyzed the personality of non-human objects such as products, stores, brands, and websites, this study analyzed the personality of Instagram accounts of public health organizations. It also extracted features at content and pixel levels from the photos uploaded on the organizations’ accounts and examined how they were related to the personality traits of the accounts. The results suggested that the personality of public health organizations can be summarized as being high in openness and agreeableness but lower in extraversion and neuroticism. Openness and agreeableness were the personality traits associated the most with the content-level features, while extraversion and neuroticism were the ones associated the most with the pixel-level features. In addition, for each of the two traits associated the most with either the content- or pixel- level features, their associations tended to be in opposite directions with one another. The personality traits, except for neuroticism, were predicted from the photo features with an acceptable level of accuracy.

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.

Information privacy behavior in the use of Facebook apps: A personality-based vulnerability assessment

The unauthorized use of personal information belonging to users of apps integrated with the Facebook platform affects millions of users. Crucially, although privacy concerns and awareness have increased, the use of these apps, and related privacy behaviors, remain largely unchanged. Given that such privacy behaviors are likely influenced by individuals' personality traits, it is imperative to better understand which personality traits make individuals more vulnerable to such unauthorized uses. We build on a recontextualized version of the theory of planned behavior (TPB) to evaluate the influence of the Big Five personality traits on attitudes toward Facebook privacy settings, social norms, and information privacy concerns (IPCs)—all within the context of Facebook app use. To evaluate this study's model, we analyzed 576 survey responses by way of partial least squares path modeling. Results indicate that highly extraverted individuals are particularly vulnerable to privacy violations (e.g., unauthorized use of personal information) because of their negative attitudes toward Facebook privacy settings. Our post hoc analysis uncovered interesting combinations of personality traits that make individuals particularly vulnerable to the unauthorized use of app-based information. In particular, the combination of extraversion and conscientiousness had a negative effect on individuals' attitude toward privacy settings. We also found a significant negative relationship between IPCs and intention to use Facebook apps. Finally, we found a positive relationship between social norms and intentions. Taken together, these results infer that individuals are likely to be influenced by their peers in the use of Facebook apps but that their intentions to use these apps declines as privacy concerns increase.