DEA-RNN A Hybrid Deep Learning Approach for Cyberbullying Detection in Twitter Social Media Platform

Abstract

Cyberbullying (CB) has become increasingly prevalent in social media platforms. With the 
popularity and widespread use of social media by individuals of all ages, it is vital to make 
social media platforms safer from cyberbullying. This paper presents a hybrid deep learning 
model, called DEA-RNN, to detect CB on Twitter social media network. The proposed DEA
RNN model combines Elman type Recurrent Neural Networks (RNN) with an optimized 
Dolphin Echolocation Algorithm (DEA) for fine tuning the Elman RNN's parameters and 
reducing training time. We evaluated DEA-RNN thoroughly utilizing a dataset of 10000 tweets 
and compared its performance to those of state-of-the-art algorithms such as Bi-directional long 
short term memory (Bi-LSTM), RNN, SVM, Multinomial Naive Bayes (MNB), Random 
Forests (RF). The experimental results show that DEA-RNN was found to be superior in all 
the scenarios. It outperformed the considered existing approaches in detecting CB on Twitter 
platform. DEA-RNN was more efficient in scenario 3, where it has achieved an average of 
90.45% accuracy, 89.52% precision, 88.98% recall, 89.25% F1-score, and 90.94% specificity...