Detection of Cyberbullying on Social Media Using Machine Learning

Detection of Cyberbullying on Social Media Using Machine Learning

Vidhya.S1, Dheivani.A2, Gopinath.S3,Manoj Kumar.S4 ,Sobiya.S 5.

Assistant Professor1,Department of Computer Science.

UG Scholar2,3,4,5, Department of Computer Science.

Angel college of engineering and technology, Tirupur -641665.

Abstract—cyberbullying was confined to boundaries and  has now moved online. Harassment is one form of cyberbullying. Social media cyberbullying incidents are escalating. Insulting words are dynamic, and the same word can have several meanings according to the context. Simply because a comment contains such a word, it cannot be classified as bullying. Hence, Labeling comments,  and keyword spotting techniques are inadequate. Other languages have addressed this issue using lexical databases such as WordNet which provides synonyms and homonyms of words. Since there is no proper lexical database developed for the English language, detecting a word as bullying is a challenge. Therefore, we used rules to overcome this issue. Twitter comments with profane words were collected, outliers were removed, and the remaining tweets were pre-processed. These rules were used for feature extraction. Afterward, we applied Support Vector Machine (SVM), K-nearest neighbor (KNN), and Naïve Bayes algorithms. The results show that SVM with an RBF kernel performs better with an F1-score of 91%. The novelty of this research is the focus on English language cyberbully detection which has not been addressed before.

Keywords—KNN, SVM, RBF, Cyberbully, Social Media, Machine learning, Personal attack, Hate speech, Feature extraction