RoBERTa reaches 93.02% accuracy on binary IMDb sentiment classification and outperforms other tested models, with a soft-voting ensemble of all models yielding further gains.
Implementation of keyword extraction using term frequency-inverse document frequency (tf-idf) in python
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From TF-IDF to Transformers: A Comparative and Ensemble Approach to Sentiment Classification
RoBERTa reaches 93.02% accuracy on binary IMDb sentiment classification and outperforms other tested models, with a soft-voting ensemble of all models yielding further gains.