Author:
Unnava Srinadh,Parasana Sankara Rao
Abstract
The popularity of online social networks has increased the prevalence of cyberbullying, making it necessary to develop efficient detection and classification methods to mitigate its negative consequences. This study offers a comprehensive comparative analysis of various machine-learning techniques to detect and classify cyberbullying. Using various datasets and platforms, this study investigates and compares the performance of various algorithms, including both conventional and cutting-edge deep learning models. To determine the best practices in various scenarios, this study includes a thorough review of feature engineering, model selection, and evaluation measures. This study also examines how feature selection and data preprocessing affect classification precision and computational effectiveness. This study provides useful information on the advantages and disadvantages of various machine learning algorithms for detecting cyberbullying through experimentation and comparative research. The results of this study can help practitioners and researchers choose the best methods for particular applications and support ongoing efforts to make the Internet safer.
Publisher
Engineering, Technology & Applied Science Research