Anticipating preferences and services for hotel guests
Machine learning can identify and analyze guests’ preferences and interests and give them tailor-made recommendations. Algorithms can organize large volumes of customer data (including biometric data) to draw conclusions about, for example, a customer’s go-to drink or room preference. Identifying these partialities can enhance the customer experience and ultimately strengthen sales.
Many of these AI systems (particularly in the biometric and emotion markets) are in the early development stages. If these tools are not appropriately developed and trained with a high-quality data set, there is a significant risk of profiling, bias and inaccuracy. Biometric data is particularly sensitive and falls under the General Data Protection Regulation’s (GDPR) definition of “special category data.” The UK Information Commissioner’s Office has warned that it will investigate organizations that fail to act responsibly when deploying biometric and emotional analysis technologies.
Dynamic and personal pricing
Historically, hotel managers set fixed-price bands for their hotels based on the city and the season. This was a time-consuming process that did not respond to surges in demand and failed to maximize revenue. Machine learning can automate this process by updating room prices in response to changes in demand, maximizing room occupancy and increasing revenue per room. It also can provide personalized pricing to different consumers based on their purchase history and inferred price elasticity.
Regulators have identified potential concerns, however, with these pricing mechanisms. For example, the UK Competition and Markets Authority has highlighted that such practices may be harmful to consumers because they may be difficult to detect, target vulnerable consumers and have unfair distributive effects.
Finding and rooting out fake social media reviews
Social media reviews are an important part of the booking experience, helping customers make purchase decisions and providing a way for businesses or platforms in the hospitality industry to build trust and credibility. To realize such benefits, it is critical that social media reviews reflect guests’ and customers’ real experiences.
In recent years, increasing numbers of fraudulent social media reviews have been appearing for travel services, hosts and other hospitality businesses, and these fake reviews can damage trust and integrity among customers. Rooting out false or fraudulent reviews can be achieved with machine learning, which detects unusual patterns in reviews by employing language processing methods.
However, current laws lack reliable definitions and legal frameworks to govern the use of AI for this purpose. In particular, the EU Commission’s proposal for an AI Act is still under review, which means that the use of AI still entails legal risks, especially if, for example, AI causes genuine reviews to be deleted.
Conclusion
Overall, great AI applications beckon for the hospitality industry, including supporting guest service and pricing and ensuring true representations appear in social media reviews. The industry does need to consider legal risks and uncertainties, but existing and proposed rules and laws are promising, and they provide a future-oriented basis for AI deployment in the hospitality sector.