metadata
languages:
- ru
multilinguality:
- monolingual
size_categories:
- 10K<n<100K
task_categories:
- text-classification
- sentiment-analysis
task_ids:
- sentiment-classification
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)