A survey on big data and machine learning in tourism and economic development

Tourism is increasingly shaped by data, and the convergence of Big Data and Machine Learning is quietly rewriting how destinations, firms, and policymakers make decisions, compete, and create value. This survey synthesizes peer-reviewed research on core applications such as demand forecasting, dynamic pricing, context-aware recommendation, and sentiment analysis across reviews and social media, while also tracing their economic implications for productivity, revenue management, and new business formation. To complement the literature, we include a structured cross-regional survey of tourism organizations that reports widespread awareness and adoption alongside uneven technological readiness. Results indicate generally positive perceived economic effects, yet they also suggest that performance gains are not guaranteed by organization size or years of technology use, which may hint that strategic alignment and data governance matter more than scale alone. We distill the principal technical and managerial challenges that limit impact in practice, including privacy and security risks under multi-jurisdictional regulation, fragmented and heterogeneous data infrastructures, skill shortages, model opacity, and the lack of standardized, open benchmarks tailored to tourism. Building on these findings, we outline a research and practice agenda that emphasizes privacy-preserving learning, interpretable and domain-specific models for real-time operations, interoperable architectures spanning IoT and streaming data, and shared evaluation datasets to enable cumulative progress. The review’s contribution is twofold: it integrates the methods and use cases of Big Data and Machine Learning in tourism with a clear view of economic outcomes, and it offers actionable priorities for responsible scaling. For academics, it maps fertile areas for rigorous, comparable studies; for industry and public agencies, it provides guidance on where investments, capability building, and governance can most plausibly translate data into resilient and inclusive tourism development.