It's a dataset consisting of news articles of SBI from time stamp 01-01-2021 to 18-09-2024.
The 'dataset.csv' file consists collection of data having higher importance in the article along with the sentiment score.
Description of the features:
- published_at : the time-series when the article was published.
- uuid : uniquely identify the article.
- sentiment_score : takes continuous values between '-1' to '+1'.
'-1' being the most negative sentiment and
'+1' being the most positive sentiment
'0' being the neutral sentiment.
- article : containing the main highlights of the article.
There were multiple highlights in some articles. Sentiment score was calculated(manually) for each of the highlights and the final score for an article is calculated by taking the average of scores for each of individual highlight score.
SBI
- Data
- sbi_csv
- metadata.csv
- entities.csv
- highlights.csv
- metadata_similar.csv
- entities_similar.csv
- highlights_similar.csv
See /SBI/Data/preview.png to know how the data looks like.