Training a Machine to See What’s Beautiful (esp. for Hotel Photos)
Training a Machine to See What’s Beautiful (esp. for Hotel Photos)


We’ve all heard that “a picture is worth a thousand words”. That definitely applies to our hotel price comparison service, a daughter company of Axel Springer. Each hotel supplies us with dozens of images, presenting us with the challenge of choosing the most “attractive” image for each pitch on our offer comparison pages since photos can be just as important for bookings as reviews. Given millions of hotel offers, we end up with more than 100 million images which require an “attractiveness” assessment.

Although teaching and training a computer to decide what makes a “beautiful” hotel photo is a hard problem, it’s not impossible. In this talk, we will present how we solved this difficult problem. In particular, we will share our training approaches and the peculiarities of the models. We will also show the “little tricks” that were key to solving this problem.


Dat is heading Axel Springer AI, the artificial intelligence unit of Axel Springer SE which is the largest digital publishing house in Europe. His goal is to make AI more accessible within Axel Springer and hence drive innovation within the group. His ultimate plan is to turn Axel Springer into an AI first company.

Dat's interests are diverse from traditional machine learning, deep learning, AI in general to computer vision and NLP. He is a regular speaker and has presented at several renowned conferences. He also blogs about his work on Medium.