This seminar, hosted by The Centre for Competitiveness of the Visitor Economy (COVE), is part of Dr Hengyun Li's IAS Fellowship.
Hotel demand and revenue prediction is central to revenue management. Although prior studies increasingly rely on online reviews to improve prediction, such human-generated signals are systematically limited by selective participation and selective expression. Meanwhile, research has paid scant attention to whether AI agents can provide complementary predictive information, especially through role-playing designs that capture heterogeneous market viewpoints, and to the mechanisms through which aggregating such information enhances predictive performance. We address these questions by developing RAAHRP, a role-playing AI-agent hotel revenue prediction model that fuses human-generated online reviews, hotel information, and AI-generated content reflecting six market roles. Using empirical analyses, we show that RAAHRP outperforms benchmark models in hotel revenue prediction. We further find that role-playing AI agents contribute differentiated and complementary predictive signals, and that the performance gains from AI-generated content are jointly driven by information scale and informational diversity, with scale exerting the stronger effect in our context. Our findings advance research on human-AI fusion by conceptualising AI agents not merely as prediction tools, but as complementary information generators. They also deepen understanding of the predictive value of role-playing AI-agent outputs and provide practical guidance for designing AI-augmented revenue management systems in hospitality.
Dr Hengyun (Neil) Li is an Associate Professor at The Hong Kong Polytechnic University and a Fellow of the Institute for Advanced Studies at the University of Surrey, UK. His research focuses on the digital and intelligent transformation of culture and tourism, the platform economy, and human-centred AI. His work has been cited approximately 7,000 times on Google Scholar. He has been recognised among the World’s Top 2% Most-Cited Scientists and was named in Research.com’s World’s Best Scientists 2026 (Business and Management). Dr. Li has published more than 30 articles in the top three journals in tourism management, with several papers ranking among the Top 8 most-cited publications over the past three years. His recent research has become increasingly interdisciplinary, with studies published in leading AI conferences such as ACM WWW and KDD, and ongoing work under revision at top journals in information systems and operations management, including ISR, POMS, and JMIS. He has received multiple honours, including recognition as one of the “50 Most Influential and Productive Scholars in Tourism Management in Greater China”, The Hong Kong Polytechnic University Faculty Award for Outstanding Young/Early Career Researcher (2025, 2021), Faculty Award for Research Excellence (2025, 2021), and selection as a University of South Carolina Breakthrough Graduate Scholar. As Principal Investigator, he has led multiple competitive projects funded by the National Natural Science Foundation of China, the Hong Kong Research Grants Council, and the Hong Kong Scholars Program.
Photo by Francesca Saraco on Unsplash
The report for this workshop is coming soon, please check back later.
Room 66 MS 03, Rik Medlik Building