Tourist preference learning method based on travel behavior and inverse reinforcement learning

In order to obtain tourists′ behavior data in scenic and learn the tourists′ fine-grained preferences, tourist preference learning method based on travel behavior and inverse reinforcement learning is proposed. First of all, through the Internet of things and mobile sensor technology to collect tourists′ behavior data such as the number of taking photos, the time of visiting in the point of scenic spots. Then, the inverse reinforcement learning algorithm is designed for the collected behavior data to perform fine-grained preference learning based on the obtained real data. The experimental results show that the method can effectively learn fine-grained preferences with a small amount of tourist behavior data based on real scenarios.