查看文章

The use of electronic communications to bully a person, usually by sending a threatening message, is called cyberbullying. Cyberbullying is abusive online behaviour that can cause very serious social problems. Twitter is the microblogging site where cyberbullying is caused very easily. People can hide their identities and bully others. They can send bad text messages and other media anonymously. In this project, machine learning methods used are random forest, support vector machine, bagging classifier, naive Bayesian, k-nearest neighbour, logistic regression, decision tree, AdaBoost and stochastic gradient descent . Moreover, these methods are analysed for detection of cyberbullying. The parameters considered for analysis are accuracy, precision, recall and F1-Score. All these algorithms are compared and the algorithm with the highest accuracy will be projected as the best solution for the above problem …