Filtered tweet_eval texts with no NRC polarity (positive / negative) lexicon words — not only the top-20 mask targets used by aieng-lab/en-sentiment-nrc. For neutral evaluation.

Usage

from datasets import load_dataset

neutral = load_dataset("aieng-lab/en-sentiment-nrc-neutral", split="train") texts = neutral["text"]

One split: train.

	Dataset Details




	


	Description

Neutral evaluation text… See the full description on the dataset page: https://huggingface.co/datasets/aieng-lab/en-sentiment-nrc-neutral.