A semi-supervised Bangla BERT pipeline classifies hyperpartisan Bangla news with a reported 95.65% accuracy, but the paper's own confusion matrix supports a lower accuracy and the evaluation is not reproducible.
Shap: A unified approach to interpreting model predictions,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Bangla BERT for Hyperpartisan News Detection: A Semi-Supervised and Explainable AI Approach
A semi-supervised Bangla BERT pipeline classifies hyperpartisan Bangla news with a reported 95.65% accuracy, but the paper's own confusion matrix supports a lower accuracy and the evaluation is not reproducible.