TripAdvisor Restaurant Reviews

Description

This dataset contains restaurant reviews from TripAdvisor for five European cities, capturing detailed information on users, restaurants (items), and reviews. It offers a comprehensive view of user experiences, opinions, and restaurant attributes.

Data Structure

User Information

userId: Unique identifier for each user (hashed).

name: Display name or username.

location: User's location (city and country).

Restaurant Information (Items)

itemId: Unique identifier for each restaurant.

name: Restaurant name.

city: City where the restaurant is located.

priceInterval: Price range.

url: Link to the restaurant’s TripAdvisor review page.

rating: Average rating score for the restaurant.

type: List of cuisine types (e.g., [Spanish, Mediterranean]).

Review Information

reviewId: Unique identifier for each review.

userId: Corresponding user who wrote the review.

itemId: Restaurant associated with the review.

title: Title of the review summarizing the user’s impression.

text: Full text of the review describing the user’s experience.

date: Date when the review was posted.

rating: Numerical score (typically from 0 to 50, where 50 represents the highest satisfaction).

language: Language of the review.

images: List of URLs pointing to images uploaded by the user (if available).

url: Link to the full review on TripAdvisor.

Code example

import pandas as pd

city = "Barcelona"
# Load restaurants
items = pd.read_pickle(f"{city}/items.pkl")
# Load users
users = pd.read_pickle(f"{city}/users.pkl")
# Load reviews
reviews = pd.read_pickle(f"{city}/reviews.pkl")