Amsterdam and Delft AI researchers team up with Booking.com to build better recommendation systems
UvA, TU Delft and Booking.com are collaborating on research into using machine learning methods to create better recommendation systems by avoiding bias, with the aim of applying these to other areas of society such as healthcare.
Building better recommendation systems
Researchers from University of Amsterdam (UvA) and Delft University of Technology (TU Delft) are teaming up with data scientists at Booking.com to improve recommendation systems.
The collaboration at the Mercury Machine Learning Lab at Amsterdam’s Innovation Center for Artificial Intelligence (ICAI) is a test of AI techniques in the real world, so that new machine learning methods can be safely developed for wider application, for example in mobility, energy and healthcare.
Currently Booking.com customers’ taps, clicks and scrolling through search results lead to recommendations based on what previous travellers have chosen and experienced. Machine learning techniques are well suited to learning such connections and preferences, especially when there is a wealth of data available.
However, these connections and preferences found in the data are not only informed by the choices other travellers have made, but also by the suggestions and selections the system showed them. This tends to skew the process, resulting in a bias in the recommendations travellers receive.
This new collaboration will aim to attempt to remove this bias, so recommendations remain accurate in a new or different context, whereby the handling of data in a safe and ethical way remains the top priority.
Joris Mooij, scientific director of the Mercury Machine Learning Lab at UvA, said: “It’s a huge opportunity for us as researchers to have access to a live dataset of global data and be able to experiment on Booking.com’s platform.”
Matthijs Spaan, one of the scientific directors from TU Delft, added: “The lab’s focus on developing better algorithms for recommender systems is highly relevant to our society as these systems guide many of our digital interactions. By addressing fundamental AI challenges, the results of the lab will also be valuable in other domains.”
Technology and talent development
The ICAI has launched several public-private collaborations aimed at technology and talent development in AI. One lab is working with the Netherlands Forensic Institute to use AI to detect deepfakes made by criminals.
Thanks to such initiatives, Amsterdam is well on its way to being a world-leader in AI and machine learning. The city is an established “living lab” for AI solutions, and programmes such as AI Technology for People ensure innovations deployed in the city truly improve citizens’ lives.