Cyberbullying detection using deep transfer learning

Abstract Social networking platforms like Facebook, Twitter, and others have numerous advantages, but they many dark sides also. One of the issues on these social is cyberbullying. The impact cyberbullying immeasurable life victims as it’s very subjective to how person would tackle this. message may be a bully for victims, it normal others. ambiguities in messages create big challenge find content. Some research has been reported address this issue with textual posts. However, image-based detection received less attention. This aims develop model that helps prevent platforms. deep learning-based convolutional neural network initially used development. Later, transfer learning models are utilized research. experimental outcomes various settings hyper-parameters confirmed better choice problem. proposed achieved satisfactory accuracy 89% best case, indicating system detects most