DETECTION OF CYBERBULLYING ON SOCIAL MEDIA USING MACHINE LEARNING AND DEEP LEARNING TECHNIQUE

Abstract

In the research of online social networks, detection of anonymous user behavior, detection of
offensive data, etc. are traditional and important research works. This project is focused on the detection of
offensive data, and bully statements in shared data of the social networks. For this system, using Machine
Learning algorithms with Text Mining concepts predicted the offensive data to get more accurate results. This
project proposed a system of “Cyber Bullying Detection (CBD) in Social Networking” for predicting bully data.
The user must be blocked if the user uses offensive words for a threshold number of times, the model should be
able to detect the words containing * marks. The model is to be built & comprise between ML (Support Vector
Machine (SVM), Random Forest (RF), Naïve Bayes (NB), and Neural Network (NN)) and DL (ANN), CNN)
This project is used two datasets, namely, ‘Hate Speech and Offensive Language Dataset’ and ‘Harassment-
Corpus Dataset’. Algorithms and calculated performance results for comparing the performance for both
datasets. To demonstrate this system designed and developed a Python-based Django web application and
showed the results