A recommender system to generate museum itineraries applying augmented reality and social-sensor mining techniques

Nowadays, museums offer technological and digital options to enrich the user experience in a visit. However, questions arise like which exhibition/museum could I visit? How to tour it and get the best experience? These questions are not easy to answer, because they do not represent tasks straightforward. Considering that the experiences of visiting a museum are now available in social networks, in which users describe, rate, and disseminate a work of art/exhibition of a museum, this information can be mined to generate tour recommendations in museums. Such recommendations could be improved by combining and applying data mining obtained from Internet of Things sensors installed in museums. In this paper, a hybrid approach to make recommendations for museum visits is proposed. It includes an Internet of Things architecture of beacons, incorporating some technologies based on semantic analysis, data mining, and machine learning. This approach integrates and combines data sources for generating and recommending indoor and outdoor itineraries for museums, which are visualized with augmented reality. The itinerary is built, taking into consideration opinions and assessments from social networks, the semantic classification of museums, and cultural activities, as well as data measured by beacon sensors in museum exhibitions. The result is a customized tour with augmented reality that contains a set of recommendations of how to visit a set of museums and obtain a better experience of the visit. A prototype of mobile application is available in the Google Play, called the “Historic Center,” with almost 500 downloads and an acceptable evaluation.