Hugging Face's logo

Datasets:
d0rj
/
geo-reviews-dataset-2023

Tasks:
Text Classification
Sentence Similarity
Token Classification
Modalities:
Text
Formats:
parquet
Languages:
Russian
Size:
100K - 1M
Tags:
reviews
yandex
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files Community
1
geo-reviews-dataset-2023 / README.md
d0rj's picture
d0rj
Update README.md
02366bc
preview code
|
raw
history blame contribute delete
2.08 kB
metadata
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
dataset_info:
  features:
    - name: address
      dtype: string
    - name: name_ru
      dtype: string
    - name: rating
      dtype: int64
    - name: rubrics
      dtype: string
    - name: text
      dtype: string
  splits:
    - name: train
      num_bytes: 369301294
      num_examples: 500000
  download_size: 170880716
  dataset_size: 369301294
license: mit
task_categories:
  - text-classification
  - sentence-similarity
  - token-classification
  - text2text-generation
language: ru
multilinguality:
  - monolingual
tags:
  - reviews
  - yandex
pretty_name: Geo Reviews Dataset 2023
size_categories:
  - 100K<n<1M
source_datasets:
  - original

Geo Reviews Dataset 2023

Yandex is making available the largest Russian-language dataset of reviews about organizations published on Yandex Maps.

Use it for academic and research purposes, share your results with us in Issues.

Dataset Description

Description

  • 500,000 unique reviews
  • Only reviews about organizations in Russia
  • Available on Yandex Maps
  • Published from January to July 2023
  • The dataset does not contain short one-word reviews
  • Reviews have been cleared of personal data (phone numbers, email addresses)

Dataset Fields

The dataset contains the following attributes:

  • Organization address (address)
  • Organization name (name_ru)
  • List of categories to which the organization belongs (rubrics)
  • User rating from 0 to 5 (rating)
  • Review text (text)

License

Distributed under MIT license.

Contacts

For any inquiries or questions regarding the dataset, please contact us at [email protected].