The advent of social media has revolutionized human communication, significantly improving individuals' lives. It makes people closer to each other, provides access to enormous real-time information, and eases marketing and business. Despite its uncountable benefits, however, we must consider some of its negative implications such as online harassment and cyberbullying. Cyberbullying is becoming a serious, large-scale problem damaging people's online lives. This phenomenon is creating a need for automated, data-driven techniques for analyzing and detecting such behaviors. In this research, we aim to address the computational challenges associated with harassment-based cyberbullying detection in social media by developing machine-learning framework that only requires weak supervision. We propose a general framework that trains an ensemble of two learners in which each learner looks at the problem from a different perspective. One learner identifies bullying incidents by examining the language content in the message; another learner considers the social structure to discover bullying. Each learner is using different body of information, and the individual learner co-train one another to come to an agreement about the bullying concept. The models estimate whether each social interaction is bullying by optimizing an objective function that maximizes the consistency between these detectors. We first developed a model we referred to as participant-vocabulary consistency, which is an ensemble of two linear language-based and user-based models. The model is trained by providing a set of seed key-phrases that are indicative of bullying language. The results were promising, demonstrating its effectiveness and usefulness in recovering known bullying words, recognizing new bullying words, and discovering users involved in cyberbullying. We have extended this co-trained ensemble approach with two complementary goals: (1) using nonlinear embeddings as model families, (2) building a fair language-based detector. For the first goal, we incorporated the efficacy of distributed representations of words and nodes such as deep, nonlinear models. We represent words and users as low-dimensional vectors of real numbers as the input to language-based and user-based classifiers, respectively. The models are trained by optimizing an objective function that balances a co-training loss with a weak-supervision loss. Our experiments on Twitter, Ask.fm, and Instagram data show that deep ensembles outperform non-deep methods for weakly supervised harassment detection. For the second goal, we geared this research toward a very important topic in any online automated harassment detection: fairness against particular targeted groups including race, gender, religion, and sexual orientations. Our goal is to decrease the sensitivity of models to language describing particular social groups. We encourage the learning algorithm to avoid discrimination in the predictions by adding an unfairness penalty term to the objective function. We quantitatively and qualitatively evaluate the effectiveness of our proposed general framework on synthetic data and data from Twitter using post-hoc, crowdsourced annotation. In summary, this dissertation introduces a weakly supervised machine learning framework for harassment-based cyberbullying detection using both messages and user roles in social media. / Doctor of Philosophy / Social media has become an inevitable part of individuals social and business lives. Its benefits, however, come with various negative consequences such as online harassment, cyberbullying, hate speech, and online trolling especially among the younger population. According to the American Academy of Child and Adolescent Psychiatry,1 victims of bullying can suffer interference to social and emotional development and even be drawn to extreme behavior such as attempted suicide. Any widespread bullying enabled by technology represents a serious social health threat. In this research, we develop automated, data-driven methods for harassment-based cyberbullying detection. The availability of tools such as these can enable technologies that reduce the harm and toxicity created by these detrimental behaviors. Our general framework is based on consistency of two detectors that co-train one another. One learner identifies bullying incidents by examining the language content in the message; another learner considers social structure to discover bullying. When designing the general framework, we address three tasks: First, we use machine learning with weak supervision, which significantly alleviates the need for human experts to perform tedious data annotation. Second, we incorporate the efficacy of distributed representations of words and nodes such as deep, nonlinear models in the framework to improve the predictive power of models. Finally, we decrease the sensitivity of the framework to language describing particular social groups including race, gender, religion, and sexual orientation. This research represents important steps toward improving technological capability for automatic cyberbullying detection.

Yes / Smart cities are shifting the presence of people from physical world to cyber world (cyberspace). Along with the facilities for societies, the troubles of physical world, such as bullying, aggression and hate speech, are also taking their presence emphatically in cyberspace. This paper aims to dig the posts of social media to identify the bullying comments containing text as well as image. In this paper, we have proposed a unified representation of text and image together to eliminate the need for separate learning modules for image and text. A single-layer Convolutional Neural Network model is used with a unified representation. The major findings of this research are that the text represented as image is a better model to encode the information. We also found that single-layer Convolutional Neural Network is giving better results with two-dimensional representation. In the current scenario, we have used three layers of text and three layers of a colour image to represent the input that gives a recall of 74% of the bullying class with one layer of Convolutional Neural Network. / Ministry of Electronics and Information Technology (MeitY), Government of India

Yes / Aggressive posts containing symbolic and offensive images, inappropriate gestures along with provocative textual comments are growing exponentially in social media with the availability of inexpensive data services. These posts have numerous negative impacts on the reader and need an immediate technical solution to filter out aggressive comments. This paper presents a model based on a Convolutional Neural Network (CNN) and Binary Particle Swarm Optimization (BPSO) to classify the social media posts containing images with associated textual comments into non-aggressive, medium-aggressive and high-aggressive classes. A dataset containing symbolic images and the corresponding textual comments was created to validate the proposed model. The framework employs a pre-trained VGG-16 to extract the image features and a three-layered CNN to extract the textual features in parallel. The hybrid feature set obtained by concatenating the image and the text features were optimized using the BPSO algorithm to extract the more relevant features. The proposed model with optimized features and Random Forest classifier achieves a weighted F1-Score of 0.74, an improvement of around 3% over unoptimized features.

Yes / Cyber-aggression is an offensive behaviour attacking people based on race, ethnicity, religion, gender, sexual orientation, and other traits. It has become a major issue plaguing the online social media. In this research, we have developed a deep learning-based model to identify different levels of aggression (direct, indirect and no aggression) in a social media post in a bilingual scenario. The model is an autoencoder built using the LSTM network and trained with non-aggressive comments only. Any aggressive comment (direct or indirect) will be regarded as an anomaly to the system and will be marked as Overtly (direct) or Covertly (indirect) aggressive comment depending on the reconstruction loss by the autoencoder. The validation results on the dataset from two popular social media sites: Facebook and Twitter with bilingual (English and Hindi) data outperformed the current state-of-the-art models with improvements of more than 11% on the test sets of the English dataset and more than 6% on the test sets of the Hindi dataset.

This study aimed to gain insight into the nature and extent of traditional and cyberbullying among Grade Six and Seven learners in four public primary schools in Benoni. Using the Social Learning Theory as a basis for understanding bullying as a learned behaviour (socially learned through the observation of authoritarian parents), a quantitative research method was applied which utilised an online self-report questionnaire to examine the relationship between bullying and the Authoritarian parenting style. Results indicate that 50.4% of learners had been victimised, while 31.6% and 8.8% had engaged in perpetrating traditional and cyberbullying, respectively at least once (N = 279). Further results revealed that the Authoritarian parenting style is significantly related to the perpetration of both types of bullying. These results bring to the fore the reciprocal relationship between both types of bullying, and indicate a need for systemic intervention at the primary school level (involving parents/caregivers). Interventions should therefore not seek to separate types of bullying into discreet problems, but rather focus on their common underlying aspects, including parenting behaviours / Psychology / M.A. (Psychology)

Children and Adolescents with emotional and behavioral disabilities (EBD) are often involved in aggression, acting out, bullying, violence, substance abuse, and juvenile crime. However, the limited Korean studies have focused primarily on bullying of students with developmental disabilities or intellectual disabilities. Therefore, the current study aimed to explore contributing factors to traditional bullying and cyberbullying in Korean children and adolescents with EBD. The current study surveyed 112 students with EBD between ages of 10 and 15 and their parents (guardians). The results revealed that internalizing problem behaviors including anxious/depression, withdrawal/depression, and somatic problems significantly affected traditional bullying victimization of Korean students with EBD. The peer support was a significant factor affecting cyberbullying victimization. Furthermore, the maternal psychological control was a meaningful factor affecting perpetration at school and in cyber world. Based on the findings, the present study described implications regarding prevention and intervention programs for addressing traditional bullying and cyberbullying victimization and perpetration.

This study aimed to gain insight into the nature and extent of traditional and cyberbullying among Grade Six and Seven learners in four public primary schools in Benoni. Using the Social Learning Theory as a basis for understanding bullying as a learned behaviour (socially learned through the observation of authoritarian parents), a quantitative research method was applied which utilised an online self-report questionnaire to examine the relationship between bullying and the Authoritarian parenting style. Results indicate that 50.4% of learners had been victimised, while 31.6% and 8.8% had engaged in perpetrating traditional and cyberbullying, respectively at least once (N = 279). Further results revealed that the Authoritarian parenting style is significantly related to the perpetration of both types of bullying. These results bring to the fore the reciprocal relationship between both types of bullying, and indicate a need for systemic intervention at the primary school level (involving parents/caregivers). Interventions should therefore not seek to separate types of bullying into discreet problems, but rather focus on their common underlying aspects, including parenting behaviours / Psychology / M. A. (Psychology)

In Tinker v. Des Moines Independent School District, the Supreme Court ruled that students have speech rights in the school environment unless the speech causes or is likely to cause 1) a substantial disruption, or 2) interferes with the rights of others. The Supreme Court has yet to hear a case involving school officials' authority to regulate electronically-delivered derogatory student speech, and no uniform standard currently exists for determining when school authorities can discipline students for such speech when it occurs off campus without violating students' First Amendment rights. The purpose of this dissertation is to examine 19 federal and state court decisions in which school authorities were sued for disciplining students for electronically delivered, derogatory speech. Eighteen of these cases involved student speech that demeaned or defamed school teachers or administrators. Only one involved speech that demeaned another student. Each case was analyzed to identify significant factors in court holdings to provide a basis for the construction of a uniform legal standard for determining when school authorities can discipline students for this type of speech. The full application of Tinker's first and second prongs will provide school officials the authority needed to address this growing problem while still protecting legitimate off-campus student cyber expression. Predictions of future court holdings and policy recommendations are included.

Traditional and cyber bullying have been identified as universal problematic issues facing adolescents, and research is needed to understand correlates associated with these phenomena. Structural equation modeling analyses examined associations between attachment to parents, traditional and cyber bullying or victimization, and mental health outcomes among 257 high school students (Average age 15.9 years). Key patterns emerged, including associations between maternal attachment and mental health outcomes; victimization and mental health concerns; and bullying and victimization in both traditional and cyber contexts. The role of attachment to mothers and fathers varied by context. Findings extend the literature by identifying risk factors in adolescence associated with bullying and victimization, as well as suggesting appropriate prevention and intervention strategies to increase adolescent well-being.

This thesis deals with cyber bullying and teachers awareness about this socio-pathological phenomenon. The theoretical part describes the concept and forms of cyber bullying, defines and describes the consequences for the cyber bullying victim. The work is dedicated to the elementary school cyber bullying prevention and defines how parents and teachers are involved with. This work describes some projects focused on safer use of the Internet. The aim of this thesis is to analyse the topic of cyber bullying and its prevention from the perspective of educators in selected primary schools. The second objective is to determine what do teachers know about the observed phenomenon and how do they perceive its prevention. Qualitative research in empirical part of the work is based on interviews with teachers. Interviews are conducted with regard to the topics content analysis in the theoretical part of the work.