A Systematic Analysis in Social Media Platforms for Cyberbullying Detecting System Using Machine Learning Techniques

Abstract

Social media platforms offer us more opportunities, than ever before and it is undeniable that they bring numerous benefits. However, it is important to confess that societies can still face humiliation, disdain, intimidation and harm from individuals or even their earls. Cyberbullying stands for the usage of technologies to defame and belittle others often through messages sent via media or email. With the growth of social media users, cyberbullying has also arisen in the method of email bullying. We undertook a project aimed at addressing this issue by examining cyberbullying through tweets using ML reinforcement algorithms such as Naive Bayes, KNN; Decision Tree, Random Forests, and Support Vector Machines (SVM). Additionally, we applied the NLTK Natural Language Toolkit techniques such as bigram analysis, trigram analysis gram analysis unigram analysis and n Naive Bayes to assess its consistency. Finally, a thorough examination was conducted on the outcomes of implementing machine learning algorithms for detecting cyberbullying while considering features and propositions. The results of our analysis demonstrate the implication of future work in identifying instances of cyberbullying.