jniimi/tripadvisor-review-rating

Dataset Overview

Dataset Name

• Name: TripAdvisor Easy Dataset
• Alias: sentiment

Dataset Features

• Feature List:
  • hotel_id: Unique hotel identifier, type int64
  • user_id: Unique user identifier, type string
  • title: Review title submitted by the user, type string
  • text: Full review body, type string
  • overall: Overall rating given by the user, type float64
  • cleanliness: Cleanliness rating, type float64
  • value: Value rating, type float64
  • location: Location rating, type float64
  • rooms: Room rating, type float64
  • sleep_quality: Sleep quality rating, type float64
  • stay_year: Year of stay, type int64
  • post_date: Review publication date, type timestamp[ns]
  • freq: Frequency, type int64
  • review: Full review, type string
  • char: Number of characters, type int64
  • lang: Language, type string

Dataset Splits

• Training Set:
  • num_examples: 201295
  • num_bytes: 368237342

Dataset Size

• Download Size: 220909380
• Dataset Size: 368237342

Dataset Configuration

• Default Configuration:
  • config_name: default
  • data_files:
    • split: train
    • path: data/train-*

Task Categories

• Task: text-classification

Language

• Language: en

Size Category

• Size: 10K<n<100K

License

• License: Apache-2.0

Intended Uses

• Direct Use: Suitable for Aspect‑based Sentiment Analysis (ABSA)
• Out‑of‑Scope Use: Follow the original data source policy

Dataset Structure

• Column Information:
  • hotel_id: Unique hotel identifier
  • user_id: Unique user identifier
  • title: Review title
  • text: Review body
  • review: Combined title + body
  • overall: Overall rating
  • cleanliness: Cleanliness rating
  • value: Value rating
  • location: Location rating
  • rooms: Room rating
  • sleep_quality: Sleep quality rating
  • date_stayed: Stay date
  • date: Review publication date