Volume - 13 | Issue-1
Volume - 13 | Issue-1
Volume - 13 | Issue-1
Volume - 13 | Issue-1
Volume - 13 | Issue-1
The advantages of social networking sites are obvious, and they have given us more possibilities than ever before. Despite the positive effects, peers, outsiders, and anonymous users frequently humiliate, abuse, bully, and torment people. The term "cyberbullying" describes the use of technology to degrade and disparage others. Hateful letters and social media posts are examples of this. Cyberbullying has surfaced as a form of bullying through social media due to the exponential rise in social media users tweets as well. The problem of harassment has been brought up by people's use of Twitter, particularly teens and young people. Lack of research exists on the characteristics influencing the spread of tweets about cyberbullying as well as the kinds of guidance and support offered in tweets for victims. 7,315 tweets related to cyberbullying were collected and examined for this research. The findings showed that tweets with specific characteristics, such as more URLs, keywords, or friends, did not always result in more retweets. The study of the tweets' emotions revealed users' attitudes towards cyberbullying to be mixed. 400 tweets were carefully selected for this study's content analysis. Tweets on harassment spanned a range of topics, from user views to current events. According to the findings, 33% of tweets included encouragement and guidance for cyberbullying sufferers. Compared to tweets discussing other aspects of harassment, these tweets received the most shares. By using ML Classification algorithms like Naive Bayes, Bi-LSTM, and BERT Classification, our research seeks to identify cyberbullying in Twitter as a potential answer to the aforementioned issue. Also, this model uses Word2Vec a technique for Natural Language Processing (NLP) for word embedding.Also,the diseases can be classified using Bert algorithm.