Balancing Privacy and Planning: Using Counterfactuals to Predict and Optimize Tourism in Wakayama City

Abstract

Leveraging predictive machine learning models on geo-spatial GPS data to analyze and predict tourist visits to Wakayama Castle, this study offers insights for urban design and policy-making. Employing counterfactual analysis, a novel approach in the tourism planning field, helps uncover key factors influencing visitation patterns and provide strategic recommendations for enhancing tourism. The comprehensive and targeted dataset of anonymized data, encompassing month long movement data, is used to train classification and regression models, which are compared and the best performing ones (Random Forest and CatBoost) are used in the counterfactual analysis. The counterfactual analysis demonstrates how changes in key factors, such as trip distance and start location, can significantly influence visitation decisions, allowing for targeted interventions to increase tourist activity. This research not only advances tourism management strategies but also prioritizes data privacy and user anonymity.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.