AI for cruisers

The cruise market has grown rapidly in recent years. It has established itself as the growth driver of tourism – worldwide. Even if external circumstances are currently causing the cruise industry to pause: It is foreseeable that the industry around Aida, TUI Cruises, MSC and Co. will continue to grow steadily. But: Where markets thrive, offers will increase. On their advertising steamer, companies are drifting further and further away from the customer and what he really wants with advertising actionism and numerous online offers. But what is that?

Afterwards, an L3S team led by Prof. Dr. Bodo Rosenhahn and the Hanoverian startup Cruisewatch will conduct research. They follow the tracks of customers who book their cruise online, using the mechanisms of AI-controlled systems. But the algorithm is laboriously feeding itself: What it needs is the “right” data – and in sufficiently large quantities. The digital cruise advisor from Cruisewatch provides this. It processes millions of data a week to provide customers in North America and Australia with dream cruises.

Apart from price and availability, it is precisely that information that is not immediately apparent that is relevant to purchasing behaviour. These are in the behavioral context of every user. The team of IT and tourism experts therefore trains the AI to interpret the context and content of the data and to learn independently with the help of neural networks. For this purpose, it tracks, for example, anonymous customer profiles of cruise ship users. After data acquisition, clustering and several trainings of the models, the matching AI is able to find the perfect cruise for the individual.

“Now it is becoming clear how important early digital transformation is,” says Titus Keuler, one of the founders of Cruisewatch. “Not only in tourism, in every industry scientifically based and data-driven analysis methods like this one are winning. They understand markets in breadth and depth almost in real time. This enables us to give our customers fact-based recommendations. And: We can react immediately and automatically to changes”.

The self-learning algorithm combines three components to better understand cruise ship customers