A Chinese Cyberbullying Detection Model based on NEZHA Model and Res-BILSTM

The traditional cyberbullying detection method based on keyword filtering has great limitations, including relying heavily on manual intervention, which has a waste of manpower and time to a certain extent. In addition, the traditional detection method cannot effectively judge the meaning of the words, due to the subjectivity and context dependence of the expression of bullying texts and some words show different meanings in different contexts. This paper proposes a Chinese cyberbullying detection method based on neural context representation model and residual network composed of BILSTM. This is a neural context representation model for Chinese understanding, which can effectively solve the above problems. The experimental results show that the proposed method has better detection effect than the comparison model on the same Chinese dataset.