About this document
Online Abuse Detection with Deep Learning by seedr evite is a document available to read on EtoBox.
The document describes research on developing deep learning models to detect online abuse and harmful comments. It discusses how online harassment is a significant problem, especially for children and teenagers. The authors experiment with convolutional and recurrent neural network models on a dataset of over 1 million annotated online comments. Their models aim to classify comments as containing abuse or not. Their approaches perform better than previous methods and analyze the nature of online conversatio
- Author
- seedr evite
- Language
- EN