Veweew/subjectivity

---
language:
- en
license: cc-by-4.0
pretty_name: Subjectivity
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
task_categories:
- text-classification
task_ids:
- multi-class-classification
---

# Subjectivity

This dataset is a Hugging Face Hub packaging of the NLTK `subjectivity` corpus,
which mirrors the Subjectivity Dataset v1.0 released by Bo Pang and Lillian Lee.

## Contents

- `train`: 10,000 text snippets
- Labels:
- `0` / `objective`: sentences from IMDb plot summaries
- `1` / `subjective`: snippets from Rotten Tomatoes reviews

## Notes

- The original corpus does not define official train/validation/test splits, so
this packaging preserves the full dataset as a single `train` split.
- Text is already lower-cased in the upstream corpus.

## Source

- NLTK data package: <https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/corpora/subjectivity.zip>
- Original dataset page: <http://www.cs.cornell.edu/people/pabo/movie-review-data/>

## Citation

```bibtex
@InProceedings{Pang+Lee:04a,
author = {Bo Pang and Lillian Lee},
title = {A Sentimental Education: Sentiment Analysis Using Subjectivity Summarization Based on Minimum Cuts},
booktitle = {Proceedings of the ACL},
year = {2004}
}
```