Cyberbully Detection from Twitter using Classifiers
J.I. Sheeba, S. Pradeep Devaneyan · CiiT international journal of artificial intelligent systems and machine learning · 2017
These days social communication networks become a part of the daily activity and the users of the social media also increased. The increasing use of social networks by their users leads to large amount of user communication data. And the popularity of social media causes cyberbullying and it became the major problem in communication through online. Cyberbullying leads to many severe problems, undesirable effect on human’s life and it also lead to suicides. In the existing system the unique information such as network, activity, user and tweet contents are extracted from Twitter. By the use of extracted information the cyberbullying words present in the tweet contents are detected using machine learning algorithms like Naive Bayes, Random Forest, Support Vector Machine and KNN. In the proposed work the rumor tweets and cyberbully tweets are detected, along with these the cyberbully words in the tweet comments also detected using Random Forest and Naive Bayes classifiers. The required information’s such as name, gender and age of the cyberbully tweeted persons are detected. By the use of twitter speech act classification features along with the machine learning classifiers, the rumor tweets are detected in this proposed work.