For Data Scientists & ML Engineers · Hotels Resorts Cruise Lines
Hotels, Resorts & Cruise Lines Data for Data Scientists
Hotels Resorts Cruise Lines data for data scientists: 15 datasets on one shelf. Every one delivered as API, files, or warehouse rows.
financial time series api for backtesting · alternative data for quantitative research · where to get training data for hotels resorts cruise lines models
API, files, or your warehouse. Daily, weekly, or hourly.
Which hospitality datasets qualify for machine-learning work?
This page ranks the 15 records the Datadory persona pack qualifies for data-science and ML work in hotels, resorts and cruise lines; six clear the relevance-3 bar and form the core shortlist, with five more at relevance 2.
Average quality across the 15 is 8.0 on the 0–10 rubric, above the 7.81 catalog-wide mean, and all six relevance-3 records score 8 or 9. The hotels-resorts-cruise-lines data hub indexes the full industry slice for readers who want every record rather than the modeling shortlist.
What can you train on without spending anything?
Two frozen benchmarks carry most supervised work. Hotel Booking Demand labels each of its 119,390 rows with is_canceled, lead_time, ADR, deposit type and arrival date across one resort and one city hotel (2015–2017), which makes it the reference corpus for cancellation classifiers and rate models.
Both are static snapshots: reproducible splits are a feature, current-market drift is not covered. Format coverage follows catalog norms — across the 1,744 datasets Datadory catalogs, CSV is offered by 734 (42.1%) and JSON by 675 (38.7%), and here eight of the 15 records serve CSV.
Which demand series feed forecasting and backtesting?
The two World Bank indicators cover the rest of the world: ST.INT.ARVL arrivals and ST.INT.RCPT.CD receipts, each roughly 265 economies with populated values concentrated 1995–2020 and empty columns from 2021 until UN Tourism publishes them. For the US market, US NTTO I-94 International Visitor Arrivals Program posts monthly Excel workbooks by country, visa type and port from 2000 to present, while UK ONS – Tourism Industry & Accommodation Services carries long-run inbound visits and spending under OGL v3.0 — note its monthly publication has been paused since March 2024.
Are the lower-relevance records worth parsing?
Four relevance-1 records serve narrow jobs. The AHLA – Hospitality Economic Impact & State Facts Dashboard exposes 52 state and 436 congressional-district rows as plain CSVs fetched without login — handy static features, but advocacy-derived figures modelled by Oxford Economics rather than survey microdata.
Straight answers
Where can I get training data for booking-cancellation prediction models?
Hotel Booking Demand (Kaggle – 119k Portuguese Hotel Bookings) is the standard choice: 119,390 bookings at one resort and one city hotel, July 2015 through August 2017, with 32 fields including is_canceled, lead_time, ADR, deposit type and stay length.
Is there a financial time series API for backtesting travel-demand models?
Yes, on quantities rather than prices. Eurostat's JSON-stat API adds regional NUTS 2 detail for European backtests.
What works as alternative data for quantitative research on hotels?
Each diverges from official statistics, but the first two are quota-bound and the last is advocacy-derived, so weight accordingly.
Rows before rollout
Sample rows from any shelf entry — the field dictionary and coverage notes ride along. If the shelf misses what you need, say so; sourcing requests are half our job.
Talk to us