A STUDY ON DEEP LEARNING ALGORITHMS FOR MULTIMODAL AND MULTILINGUAL CYBERBULLYING DETECTION

2021 ◽  
pp. 50-54
Author(s):  
Vijayakumar V ◽  
Hari Prasad D
Keyword(s):  
Deep Learning ◽  
Text Messages ◽  
Cyber Bullying ◽  
Physical And Mental Health ◽  
Modal Data ◽  
Learning Techniques ◽  
The Social ◽  
Social Media Platforms ◽  
Cyberbullying Detection ◽  
Multiple Languages

With the increased utilization of the internet and social media platforms, can foster destructive or harmful behaviors such as cyberbullying. Cyberbullying poses signicant threat to physical and mental health of the victims. There is a demand for automatic detection and prevention of cyberbullying. In Social networks, there is a big challenge to detect the cyber bullying event and to control all the cyberbullying content and languages that users post. Due to complexity of multiple languages and cross-mix languages used in cyberbullying, the detection has remained only mildly satisfying. And also recently, images and videos dominate the social feeds in addition to text messages and comments. Machine learning and deep learning techniques can be helpful to detect the bullies and can generate a model to automatically detect multi-lingual cyberbullying actions. Deep neural architectures are useful to model, learn and fuse multi-modal data for cyber bullying detection. This paper proposes a detailed review on machine and deep learning approach for detecting and preventing multimodal and multilingual cyberbullying.

scholarly journals Proactively Discouraging Cyberbullying Activities

2021 ◽  
Vol 9 (10) ◽  
pp. 1601-1607
Author(s):  
Puneetha KR
Keyword(s):  
Social Media ◽  
Language Processing ◽  
Leisure Activities ◽  
Cyber Bullying ◽  
Future Research ◽  
Internet Users ◽  
The Social ◽  
Social Media Platforms ◽  
Root Word ◽  
Cyberbullying Detection

Abstract: Research into cyberbullying detection has increased in recent years, due in part to the proliferation of cyberbullying across social media and its detrimental effect on young people. Cyber bullying is one of the most common problems faced by the internet users making internet a vulnerable space hence there has to be some detection that is needed on the social media platforms. Detecting the bullies online at the earliest makes sure that these platforms are safer for the user and internet indeed becomes a platform to share information and use it for other leisure activities. Even though there has been some research going on implementing detection and prevention of cyber bullying, it is not completely feasible due to certain limitations imposed. In this paper lexicon-based approach of the NLTK sentiwordnetis used to differentiate the positive and negative words and produce results. These words are given negative and positive values greater than or less than zero for positive and negative words respectively. Lexicon based systems utilize word lists and use the presence of words within the lists to detect cyberbullying. Lemmatization is used to find the root word. This paper essentially maps out the state-of-the-art in cyberbullying detection research and serves as a resource for researchers to determine where to best direct their future research efforts in thisfield. Keywords: Abuse and crime involving computers, natural language processing, sentiment analysis, social networking

scholarly journals Online Public Shaming Approach using Deep Learning Techniques

2021 ◽  
Vol 23 (3) ◽  
Author(s):  
Mehdi Surani ◽  
◽  
Ramchandra Mangrulkar ◽  
Keyword(s):  
Deep Learning ◽  
Hate Speech ◽  
Severe Depression ◽  
Learning System ◽  
Online Bullying ◽  
Digital World ◽  
Learning Techniques ◽  
Proposed Model ◽  
Social Media Platforms ◽  
The Impact

Public shaming on social media platforms like Twitter / Instagram / Facebook etc. have recently increased from the past years. This results in affecting an individual’s social, political, mental and financial life. The impact can range from mild bullying to severe depression. With the growing leniency on these social platforms, many people have started misusing the opportunity by turning to online bullying and hate speech. When something is posted online, it stays there forever and it becomes extremely hard taking something out of the digital world. Manually locating and categorizing such comments is a lengthy procedure and just cannot be relied upon. To solve this challenge, automation was performed to identify and classify the shamers. This has been done using the classic SVM model which worked on a given quantity of data. To identify the negative content being posted and discussed online, this paper further explores the deep learning system which can successfully classify these content pieces into proper labels. The text-based Convolution Neural Network (CNN) is the proposed model in this paper for this analysis.

scholarly journals Link Prediction in Facebook using Web Scrapping and Deep Learning Techniques

2021 ◽  
Vol 10 (1) ◽  
pp. 194-197
Keyword(s):  
Social Networks ◽  
Deep Learning ◽  
Link Prediction ◽  
Machine Learning Techniques ◽  
Feature Engineering ◽  
Internet Technologies ◽  
Learning Techniques ◽  
The Social ◽  
Feature Based ◽  
Day By Day

As Internet technologies develop continuously social networks are getting more popular day by day. People are connected with each other via virtual applications. Using the Link Prediction in social networks more people get connected, may be they are friends, may be work together at the same workplace and may be their education are. Machine learning techniques are used to analyze the link between the nodes of the network and also create a better link prediction model through deep learning. The objective of this research is to measure the performance using the different techniques to predict link between the social networks. Using deep learning, feature engineering can be reduced for link prediction. In this research, the feature based learning is used to predict the link for better performance. Dataset is obtained by scraping the profile of Facebook users and they are used along with the random forest and graph convolution neural network to measure the performance of link prediction in social networks.

scholarly journals Comparative Analysis of Deep Learning Techniques to detect Online Public Shaming

2021 ◽  
Vol 40 ◽  
pp. 03030
Author(s):  
Mehdi Surani ◽  
Ramchandra Mangrulkar
Keyword(s):  
Social Media ◽  
Deep Learning ◽  
Freedom Of Speech ◽  
Learning Approaches ◽  
Learning Models ◽  
Media Usage ◽  
The Past ◽  
Learning Techniques ◽  
Social Media Platforms ◽  
Machine Learning Models

Over the past years the exponential growth of social media usage has given the power to every individual to share their opinions freely. This has led to numerous threats allowing users to exploit their freedom of speech, thus spreading hateful comments, using abusive language, carrying out personal attacks, and sometimes even to the extent of cyberbullying. However, determining abusive content is not a difficult task and many social media platforms have solutions available already but at the same time, many are searching for more efficient ways and solutions to overcome this issue. Traditional models explore machine learning models to identify negative content posted on social media. Shaming categories are explored, and content is put in place according to the label. Such categorization is easy to detect as the contextual language used is direct. However, the use of irony to mock or convey contempt is also a part of public shaming and must be considered while categorizing the shaming labels. In this research paper, various shaming types, namely toxic, severe toxic, obscene, threat, insult, identity hate, and sarcasm are predicted using deep learning approaches like CNN and LSTM. These models have been studied along with traditional models to determine which model gives the most accurate results.

Co’FeeL: An Insightful COVID’19 Emotion Analysis based on Worldwide Twitter Data and Advanced Deep Learning Techniques (Preprint)

2020 ◽  
Author(s):  
Shruti Patil ◽  
Nikhil Matta ◽  
Dr.Ketan Kotecha
Keyword(s):  
Deep Learning ◽  
Management Strategies ◽  
Human Beings ◽  
Emotional Wellbeing ◽  
Social Fabric ◽  
The Public ◽  
Learning Techniques ◽  
Social Media Platforms ◽  
The Government ◽  
The Masses

UNSTRUCTURED The coronavirus outbreak has altered the complete living pattern of human beings across the globe. It has disrupted the way we live, work, and play. The preventive norm of social distancing has created a void in our society’s social fabric. It has affected us not only physically or financially but has created a greater impact on our emotional wellbeing as well. This distress in our emotional quotient is a result of multiple factors such as financial implications, family member’s behavior, and support, country-specific lockdown protocols, media influence, fear of a pandemic, etc. For efficient pandemic management, there is an urgent need to understand these emotional variations among the masses, as they would provide great insights regarding the public sentiments towards the government pandemic management policies. It will also bring to light the weaker emotional sections of the masses, which we should be strengthening to face further situations more effectively. During this time, more and more people have used various social media platforms such as Twitter to stay connected and express their feelings and concerns. In this paper, we have collected and analyzed over 1 million tweets of the last five months (February-June 2020) using advanced deep learning techniques such as Transfer Learning and Robustly Optimized BERT Pretraining Approach (RoBERTa) to study countrywide variations in emotions. We have categorized emotions into eight classes viz. anger, depression, enthusiasm, hate, relief, sadness, surprise, and worry. The outcome of this analysis, which is represented in the form of graphs, provides insights into how emotions have changed over time for various countries. These insights can be very useful not only in formulating effective pandemic management strategies but also to devise predictive strategies for the emotional wellbeing of the country as a whole and citizens in particular for future distress events.

scholarly journals MaLang: A Decentralized Deep Learning Approach for Detecting Abusive Textual Content

2021 ◽  
Vol 11 (18) ◽  
pp. 8701
Author(s):  
Pranav Kompally ◽  
Sibi Chakkaravarthy Sethuraman ◽  
Steven Walczak ◽  
Samuel Johnson ◽  
Meenalosini Vimal Cruz
Keyword(s):  
Deep Learning ◽  
Hate Speech ◽  
System Level ◽  
Learning Approach ◽  
Unique Identifier ◽  
Detection Systems ◽  
Learning Techniques ◽  
Cyberbullying Detection ◽  
Textual Content ◽  
Deep Learning Model

Cyberbullying is a growing and significant problem in today’s workplace. Existing automated cyberbullying detection solutions rely on machine learning and deep learning techniques. It is proven that the deep learning-based approaches produce better accuracy for text-based classification than other existing approaches. A novel decentralized deep learning approach called MaLang is developed to detect abusive textual content. MaLang is deployed at two levels in a network: (1) the System Level and (2) the Cloud Level, to tackle the usage of toxic or abusive content on any messaging application within a company’s networks. The system-level module consists of a simple deep learning model called CASE that reads the user’s messaging data and classifies them into abusive and non-abusive categories, without sending any raw or readable data to the cloud. Identified abusive messages are sent to the cloud module with a unique identifier to keep user profiles hidden. The cloud module, called KIPP, utilizes deep learning to determine the probability of a message containing different categories of toxic content, such as: ‘Toxic’, ‘Insult’, ‘Threat’, or ‘Hate Speech’. MaLang achieves a 98.2% classification accuracy that outperforms other current cyberbullying detection systems.

scholarly journals COVID-19 Related Sentiment Analysis Using State-of-the-Art Machine Learning and Deep Learning Techniques

2022 ◽  
Vol 9 ◽  
Author(s):  
Zunera Jalil ◽  
Ahmed Abbasi ◽  
Abdul Rehman Javed ◽  
Muhammad Badruddin Khan ◽  
Mozaherul Hoque Abul Hasanat ◽  
...  
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Deep Learning ◽  
Sentiment Analysis ◽  
Social Networking Sites ◽  
Daily Lives ◽  
The Public ◽  
Learning Techniques ◽  
Social Media Platforms ◽  
Twitter Users

The coronavirus disease 2019 (COVID-19) pandemic has influenced the everyday life of people around the globe. In general and during lockdown phases, people worldwide use social media network to state their viewpoints and general feelings concerning the pandemic that has hampered their daily lives. Twitter is one of the most commonly used social media platforms, and it showed a massive increase in tweets related to coronavirus, including positive, negative, and neutral tweets, in a minimal period. The researchers move toward the sentiment analysis and analyze the various emotions of the public toward COVID-19 due to the diverse nature of tweets. Meanwhile, people have expressed their feelings regarding the vaccinations' safety and effectiveness on social networking sites such as Twitter. As an advanced step, in this paper, our proposed approach analyzes COVID-19 by focusing on Twitter users who share their opinions on this social media networking site. The proposed approach analyzes collected tweets' sentiments for sentiment classification using various feature sets and classifiers. The early detection of COVID-19 sentiments from collected tweets allow for a better understanding and handling of the pandemic. Tweets are categorized into positive, negative, and neutral sentiment classes. We evaluate the performance of machine learning (ML) and deep learning (DL) classifiers using evaluation metrics (i.e., accuracy, precision, recall, and F1-score). Experiments prove that the proposed approach provides better accuracy of 96.66, 95.22, 94.33, and 93.88% for COVISenti, COVIDSenti_A, COVIDSenti_B, and COVIDSenti_C, respectively, compared to all other methods used in this study as well as compared to the existing approaches and traditional ML and DL algorithms.

A Review on Natural Disaster Detection in Social Media and Satellite Imagery Using Machine Learning and Deep Learning

2021 ◽  
Author(s):  
Swapandeep Kaur ◽  
Sheifali Gupta ◽  
Swati Singh ◽  
Tanvi Arora
Keyword(s):  
Machine Learning ◽  
Social Media ◽  
Deep Learning ◽  
Natural Disaster ◽  
Satellite Imagery ◽  
Human Life ◽  
High Volume ◽  
Volume Data ◽  
The Social ◽  
Social Media Platforms

A disaster is a devastating incident that causes a serious disruption of the functions of a community. It leads to loss of human life and environmental and financial losses. Natural disasters cause damage and privation that could last for months and even years. Immediate steps need to be taken and social media platforms like Twitter help to provide relief to the affected public. However, it is difficult to analyze high-volume data obtained from social media posts. Therefore, the efficiency and accuracy of useful data extracted from the enormous posts related to disaster are low. Satellite imagery is gaining popularity because of its ability to cover large temporal and spatial areas. But, both the social media and satellite imagery require the use of automated methods to avoid the errors caused by humans. Deep learning and machine learning have become extremely popular for text and image classification tasks. In this paper, a review has been done on natural disaster detection through information obtained from social media and satellite images using deep learning and machine learning.

scholarly journals Governing the Risks of Cyber Bullying in the Workplace During the Era of Covid-19

2021 ◽  
Vol 6 (10) ◽  
pp. 334-342
Author(s):  
Wan Rosalili Wan Rosli ◽  
Syazni Nadzirah Ya’cob ◽  
Mardiyah Hayati Abu Bakar ◽  
Mimi Sintia Mohd Bajury
Keyword(s):  
Mental Health Problems ◽  
Secondary Data ◽  
Legal Framework ◽  
Cyber Bullying ◽  
Negative Consequences ◽  
Physical And Mental Health ◽  
Post Traumatic Stress ◽  
Social Media Platforms ◽  
Post Traumatic ◽  
Online Sources

With the advancement of ICT, cyberbullying has become more common than ever before, particularly in modern workplaces. With the requirement of working from home during the pandemic, cyberbullying within the workplace has skyrocketed within the past year. Cyberbullying can be classified as a traditional crime that has transcended to cyberspace as a result of technological advancements and the proliferation of numerous social media platforms. Despite widespread public concern about such crime in Malaysia, the legislative response to this crime is still somewhat slow due to the gaps in the current legislation governing cyberbullying. The legal landscape governing cyberbullying is still insufficient, due to the current legal framework being too general, making investigation and prosecution of the crime difficult. Cyberbullying can result in Post-Traumatic Stress Disorder, psychological problems, major physical and mental health problems, and even suicide. The purpose of this article is to investigate the notions of cyber bullying harassment, the risks associated with such crimes, and the legal and management mechanisms for dealing with such crimes. This research makes use of a doctrinal content analysis as well as secondary data from the law, academic journals, books, and online sources. According to the authors, unequal power relations in the workplace, anonymity, and cross-border connectedness are some of the rationales for cyberbullying, which can be expressed in a variety of ways with negative consequences for employers and employees alike. The inadequacy of the present traditional and computer-specific legislation in dealing with such crime necessitates the management of such crime.

scholarly journals A Multichannel Deep Learning Framework for Cyberbullying Detection on Social Media

2021 ◽  
Author(s):  
Munif Alotaibi ◽  
Bandar Alotaibi ◽  
Abdul Razaque
Keyword(s):  
Social Networks ◽  
Social Media ◽  
Deep Learning ◽  
Language Processing ◽  
Online Social Networks ◽  
Hate Speech ◽  
Message Service ◽  
Integral Role ◽  
Social Media Platforms ◽  
Cyberbullying Detection

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%.

Sign in / Sign up

Export Citation Format

Share Document