datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
text-classification-checkpoint-downloadstext-classification-comparison
Text Classification Comparison: Supervised vs Unsupervised on stanfordnlp/imdb
Dataset
stanfordnlp/imdb (Maas et al., 2011)
50,000 IMDB movie reviews: 25,000 train / 25,000 test
Binary sentiment: 0 (negative) / 1 (positive), perfectly balanced in both splits
Preprocessing: TF-IDF (15,000 features, bigrams, sublinear TF, min_df=3)
Models Chosen
Model
Type
Key Reference
Why
Logistic Regression
Linear supervised
McFadden (1974); Ng &… See the full description on the dataset page: https://huggingface.co/datasets/IntimateUser6969/text-classification-comparison.text-classificationtext_classification_test