Enhancing Cyber Bullying Detection Using Convolutional Neural Network

• Published in: 2023 4th International Conference on Smart Electronics and Communication (ICOSEC) pp. 1260 - 1267
• Main Authors: Dedeepya, P., Karishma, D., Manuri, Siva Ganesh, Raghuvaran, Tata, Shariff, Vahiduddin, Sindhura, S
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20.09.2023
• Subjects: Analytical models; Blogs; Classifiers; Convolutional neural networks; Cyber bullying; Cyberbullying; Deep learning; Legislation; Natural Language Processing; Software; Twitter API
• DOI: 10.1109/ICOSEC58147.2023.10276007

The abstract focuses on the prevalence of cyber bullying among young individuals on social media platforms. As the popularity of these platforms continues to grow, the prevalence of cyber bullying incidents is also increasing. The aim is to leverage Deep learning methods and create a DL model capable of automatically identifying social media bullying through the analysis of word similarities within tweets. To find word similarities in the tweets that were transcribed by bullies, use deep learning, to create a deep learning model that can mechanically detect bullying acts on social media. The purpose of the work is to demonstrate the creation of software which can recognize harassed tweets, posts, and other forms of online communication. It has been recommended for practice a model constructed on deep learning to find bullying on Twitter. CNN is preferred to understand the significance of text and to group it appropriately, whilst NLP is developed for the purpose of processing the records. Moreover, the Twitter API is utilized in order to get tweets, and then those tweets are fed into a model which concludes whether or not they establish bullying.