Machine Learning and Affect Analysis Against Cyber-Bullying

IJCSIS Vol. 18 No. 10 OCT 2020, 2020

Abstract—On an average, 20% to 40% of all teenagers have been mistreated online. With appropriate detection of possible harmful messages, successful prevention can be achieved. However, there is a requirement for intelligent systems to identify possible risks automatically, given how the Web is overloaded with massive information. This is what encouraged us to control bullying by detecting it on different social media sites so that the people out there can take initiatives to end it. This paper surveys several attempts to use machine learning technology to detect instances of cyber bullying and summarizes their results. Overall, the results are promising, but with these results still have room for improvement Keywords-component; cyber bullying; machine learning

IJRASET, 2021

Cyber bullying has been increasing at an unfathomable rate with the rise of the internet and the increasing radius of the social networks. This increasing amount of cyber bullying serves to be a threat to many people who are an active part of these social networks, especially the people using the social media websites for posting information and communication. These crimes have affected many people and have also resulted in suicides. Thus we need an efficient system to detect these bullying attacks and reduce their amount. In the past machine learning models have been constructed to address this problem, but those models have not been very successful. In this paper we will discuss the latest technologies like Naive Bayes, SVM, BERT models, CNN and LSTM that have been applied in this area and compare their results.

Opinion mining or sentiment analysis is considered as an important application of NLP (Natural Language Processing). Opinion mining is extracting the views that people express online. Those Websites which permits social interaction and collaboration can be considered as social media site, including networking sites such as Facebook, MySpace, and Twitter. Such sites offer today's youth a platform for amusement, entertainment, thrill, correspondence and communication with friends and furthermore have developed radically and exponentially as of late. This is the reason, there are various side effects, as cyberbullying has emerged as a serious issue afflicting children, adolescents and young adults. Machine learning techniques have conceivable ability to make automatic detection of bullying messages in social media, and this could develop a healthy and comparatively safe social media environment. Social media getting more and more popular in our day today life. By the popularity of the social media affects the people who involving into it. This makes the technology to work or to feel smarter and makes us lazier. On resulting to this robust and discriminative numerical representation learning of text messages is a critical issue. Hence here we propose a learning method to tackle this issue which is named as Semantic Enhanced Marginalized Denoising Auto Encoder (smsda).

International Journal for Research in Applied Science & Engineering Technology (IJRASET), 2022

With the upward thrust of the Internet, using social media has exploded, and it's emerged because the most powerful networking platform of the ordinal century. However, exaggerated social networking oftentimes has negative consequences for society, causative to some unwanted phenomena at the side of online abuse, harassment, cyberbullying, cybercrime, and trolling. Cyberbullying causes severe mental and physical distress in several people, particularly ladies and children, and might even cause suicide damaging social impact of online harassment attracts attention. Several incidences of online harassment, equivalent to sharing personal chats, spreading rumours, and creating sexual remarks, have recently occurred everywhere on the planet. As a result, specialists are paying nearer interest to detect bullying the big texts or messages on social media. By combining natural language processing and machine learning the aim of this observation is to create and construct a powerful method for detecting online abusive and bullying texts. The accuracy stage of six different machine learning techniques is evaluated the usage by of extraordinary features, particularly the count vectorizer.

Now a day's social media have attracted a lot of web user's attention towards it for knowing the recent updates, share the new thoughts and exchange the latest updates throughout the world from one location to other location. As this is becoming more popular still it faces some common limitations in terms of controlling the abused or vulgar conversations from individual user account. In general cyber bullying is defined as the process of threatening a small kid or preteen by another child using the internet with some bad conversations and tries to make them feel nervous with these conversations. This is becoming a very severe problem in the current social media by afflicting the children, young adults with these rude messages. Hence we try to construct a machine learning approach for detecting the set of abused words in the conversations and try to block such conversations not to be spread to others. By using this proposed approach we can able to create a positive and safe communication in social media.

Cyberbullying is the use of technology as a medium to bully someone. Although it has been an issue for many years, the recognition of its impact on young people has recently increased. Social networking sites provide a fertile medium for bullies, and teens and young adults who use these sites are vulnerable to attacks. Through machine learning, we can detect language patterns used by bullies and their victims, and develop rules to automatically detect cyberbullying content.

IRJET, 2020

In today's life, the effect of social media is increasing popularly .cyber bullying has emerged as a critical afflicting children, youngster's and teenagers. System gaining knowledge of strategies make automatic detection of bullying messages in social media possible, and this may assist to assemble a healthful and safe social media environment. One critical issue in this significant research area is the reliable and discriminative processing of text messages in numerical representation. In this cyber bullying is harassment or bullying executed through digital device like computers, laptops, Smartphone and tablets. Cyber bullying can be defined as belligerent, intentional actions performed by an individual or a group of people via digital communication methods such as sending messages and posting comments against a victim. This paper, review Many companies will used this application for chatting, email notification, group chatting and meeting also. This web is helpful for people who is work on company because many people use the abusing word or comment on post that time ,this application doesn't show abused word only show there related symbols like asterisk, small square box etc. In this web, when person start chatting with another person or comment on post that time they cannot use abused words.

International Journal of Advanced Computer Science and Applications

Unless and until our society recognizes cyber bullying for what it is, the suffering of thousands of silent victims will continue." ~ Anna Maria Chavez. There had been series of research on cyber bullying which are unable to provide reliable solution to cyber bullying. In this research work, we were able to provide a permanent solution to this by developing a model capable of detecting and intercepting bullying incoming and outgoing messages with 92% accuracy. We also developed a chatbot automation messaging system to test our model leading to the development of Artificial Intelligence powered anti-cyber bullying system using machine learning algorithm of Multinomial Naïve Bayes (MNB) and optimized linear Support Vector Machine (SVM). Our model is able to detect and intercept bullying outgoing and incoming bullying messages and take immediate action.

Applied Sciences

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...

International Journal for Research in Applied Science & Engineering Technology, 2021

Cyberbullying could be an upsetting on-line wrongdoing with disturbing consequences. It seems in several forms, and in most of the social networks, it's in text format. Automatic detection of such incidents needs intelligent systems. Most of the prevailing studies have approached this drawback with typical machine learning models and therefore the majority of the developed models in these studies are applicable to one social network at a time. In recent studies, deep learning primarily based models have found their means within the detection of cyberbullying incidents, claiming that they will overcome the restrictions of the standard models, and improve the detection performance. Cyberbullying is that the use of technology as a medium to bully somebody. Though it's been a difficulty for several years, the popularity of its impact on teenagers has recently inflated. Social networking sites offer a fertile medium for bullies, and youths and young adults UN agency use these sites are susceptible to attacks. Through machine learning, we are going to realize language patterns utilised by bullies and their victims, and develop rules to automatically realize cyberbullying content. We find that our approach is with success able to determine vital variations between cyberbullying and regular media sessions, and supply a performance increase in cyberbullying detection. This paves the means for a lot of nuanced work on the utilization of temporal modelling to find and mitigate the incidence of cyberbullying. I. INTRODUCTION Day to day life is entirely or in a way dependent on the advent of internet. Thus, cyberbullying has been a major worry. With the advancement in technology, the internet has been a safe and secure sphere of communication, though the arena of social media has been prone to cybercrimes. Since the social lifestyle surpass the physical barrier of human interaction and affords inappropriate interaction with unknown people, it is important to analyse and study the domain of cyberbullying. Moreover, a well-specified law framework for cyberbullying has not been enforced in majority of the countries, therefore the data to defend the matter is unsure. Cyberbullying are often outlined because the use of a web communication to bully or harass an individual World Health Organization doesn't have potential to react [1], usually by causation messages of a threating or discouraging nature. It is evident that around 87 percent of the today's youth have witnessed some form of cyberbullying [2]. One of the rare reports [3] on cyberbullying states that 60 percent of Gulf Countries' youth overtly admit the presence of cyberbullying amongst their peers. This study conjointly states that solely quarter the predators on-line do bully their victims offline. This means that net have impressed three quarters of the predators to bully others, whereas they wouldn't have thought-about bullying physically. Cyberbullying can take various forms like Sexual Harassment, Hostile Environment, Racism, Revenge, and Retaliation. Since the offender is hidden to the victim, the problem statement gets complex. Effects of cyberbullying can range from temporary anxiety to suicide [6]. This is the reason cyberbullying is an interesting field of research. The adverse result of a cybercrime are often drastic-Cyberbullying was powerfully connected dangerous thinking compared with ancient bullying (JAMA pediatrics, 2014) [4], Hence, the requirement for a good system to spot cyberbullying and relieve the plight of distressed users. Since, cyberbullying can take place without the direct confrontation of the perpetrator, it is lot more vulnerable. Moreover, the most vicious state of bullying is that it can take place across social networks which were previously unreachable. Thus, with the proliferation of social media and web access, the act of cyberbullying too has inflated manifold. Twitter is one amongst the foremost lauded and widespread social media existing. It allows users to send and browse 140-character message. It is astonishing that about 330 million active users access the platform and nearly 500 million tweets are exchanged a day. Since about 80 percent of the user's access with their mobile phones, it has been an arena of real-time communication. A study determined that Twitter is turning into a cyberbullying playground (Xuetal, 2012). In this analysis, we tend to tend to utilize this important information and knowledge within the kind of tweets to enhance the prevailing cyberbullying detection performance. Since, Twitter is very user-friendly it enables the use of extended features like network, activity, user and tweet content, to train our detection model and improve its performance.