Advanced Cyberbullying Detection: A Hybrid Model Integrated with Naïve Bayes
M. Priyadharshini, J Hemagowri, A. Faritha Banu, B. Nithya, K.G. Srinidhi, Veerasamy Murugesh · 2024
The problem of cyberbullying is common on social media and online platforms, but it requires more advanced detection techniques. The below-captioned article suggests a holistic method to recognize cyberbullying utilizing Naive Bayes algorithm and TF-IDF (Term Frequency-Inverse Document frequency) algorithm, for text analysis of various types of social media content. Including the Naive Bayes algorithm, along with Text Preprocessor (TFIDF), for preprocessing text data drastically improves how well and quickly it is able to classify content as non-bullying or bullying. This work not only emphasize on identifying cyberbullying in social media, using various categories of data and approaches but also stresses that integrating different analysis methodologies is necessary due to the challenge nature of detecting CYB in existing multi-type social medias. Our novel model performs best in detection of cyberbullying, which can result to private and reliable online communications as well as more efficient monitoring tool for social media platforms.