Hedonic price models, social media data and AI - An application to the AIRBNB sector in us cities

OpenAlex İngilizce

The Airbnb sector has experienced exponential growth over the past decade and has led to extensive research in fields such as hospitality sciences, urban geography, tourism economics, and information management. This paper contributes to quantitative research in the Airbnb sector by focusing on the integration of digital platform data at the neighborhood level. It explores innovative methodologies for analyzing urban attractiveness by combining insights from hedonic pricing models with large-scale digital data sourced through AI-based approaches. This novel framework compares user-based valuations of accommodations derived from hedonic pricing with subjective, AI-generated neighborhood descriptions, offering new perspectives on data quality and reliability in information systems. The study also critically examines the challenges of integrating AI-generated content in information science, referencing also ‘Garbage-in Garbage-out’ and ‘Bullshit-in Bullshit-out’ concepts. Employing a multi-scalar modeling approach, the research examines Airbnb pricing dynamics across several U.S. cities, starting with Manhattan (USA) as an illustrative case. A subsequent large-scale application to additional metropolitan areas utilizes a combination of hedonic price modeling, social media data, and AI-generated urban descriptions, including a Shapley decomposition analysis. This interdisciplinary integration provides actionable insights into neighborhood attractiveness and pricing mechanisms, while highlighting methodological and empirical contributions to the broader field of information management. By employing the relationship between AI-driven textual data and quantitative modeling, this research provides added value in analyzing urban information systems and their application to digital platforms. • Using Ai for variable generation is novel. Generated variables contribute substantially to the explanatory power of the model. • By combining hedonic pricing with AI and big data analytics, the paper reflects methodological progression. • In a hedonic price method, we use Ai generated output to see if Ai generated variables can be employed as regressors.