The viewer is disabled because this dataset repo requires arbitrary Python code execution. Please consider removing the loading script and relying on automated data support (you can use convert_to_parquet from the datasets library). If this is not possible, please open a discussion for direct help.

Dataset Summary

The dataset contains user reviews about medical facilities.

In total it contains 70,597 reviews. The detailed distribution on sentiment scale is:

  • 41,419 positive reviews;
  • 29,178 negative reviews.

Data Fields

Each sample contains the following fields:

  • review_id;
  • category category of medical facility (one of 48);
  • title: review title;
  • content: review text;
  • sentiment: sentiment (positive or negative);
  • source_url.

Python

import pandas as pd
df = pd.read_json('healthcare_facilities_reviews.jsonl', lines=True)
df.sample(5)
Downloads last month
108