Transformer-based probabilistic forecasting of daily hotel demand using web search behavior

Open Webpage

Cristof Rojas, Adam Jatowt

Published: 2025, Last Modified: 05 Feb 2025Knowl. Based Syst. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Advanced ML (TFT) greatly outperforms traditional forecasting for hotel demand.•Using Google Trends data and other variables boosts forecast accuracy.•Probabilistic forecasts clarify demand uncertainty, aiding decisions.