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 restaurants. In total it contains 47,139 reviews. A review tagged with the general sentiment and sentiments on 3 aspects: food, interior, service.
Data Fields
Each sample contains the following fields:
- review_id;
- general;
- food;
- interior;
- service;
- text review text.
Python
import pandas as pd
df = pd.read_json('restaurants_reviews.jsonl', lines=True)
df.sample(5)