Datasetpublic

prithvi1029/sentiment-analysis-for-financial-news

Sentiment Analysis for Financial News (FinancialPhraseBank-style) This dataset contains financial news headlines labeled with sentiment from the perspective of a retail investor. Columns sentiment: one of negative, neutral, positive news_headline: financial news headline text Source / Context The original dataset is commonly referenced as FinancialPhraseBank and is widely used for financial sentiment benchmarking. Usage from… See the full description on the dataset page: https://huggingface.co/datasets/prithvi1029/sentiment-analysis-for-financial-news.

sourceHugging Faceapache-2.0updated 9mo agoView on Hugging Face

1likes43downloads

Dataset Card

Sentiment Analysis for Financial News (FinancialPhraseBank-style)

This dataset contains financial news headlines labeled with sentiment from the perspective of a retail investor.

Columns

—sentiment: one of negative, neutral, positive
—news_headline: financial news headline text

Source / Context

The original dataset is commonly referenced as FinancialPhraseBank and is widely used for financial sentiment benchmarking.

Usage

python

from datasets import load_dataset

ds = load_dataset("prithvi1029/sentiment-analysis-for-financial-news")
print(ds)
print(ds['train'][0])

Notes

—This repo is intended for research and educational use.

Citation

If you use this dataset, please cite the original work:

Malo, P., Sinha, A., Takala, P., Korhonen, P., & Wallenius, J. (2014). Good debt or bad debt: Detecting semantic orientations in economic texts. Journal of the American Society for Information Science and Technology.