Tourist Recommandation

The tourism industry constitutes a significant part of the revenues of the State of Israel. In 2019, the total income from 4.5 million tourists was about 21 billion NIS. To create an unforgettable experience for tourists who come to Israel, we built models that can predict the location of the next activity place (city) and the next activity itself for a tourist. We worked on tourist activity data collected by the MOT (Ministry of Tourism), the algorithm learns the tourist's details and the previous activities he did and gives a prediction for the next activity place. In addition, we have built an LSTM network that provides predictions based on each tourist's previous series of activities only. Later, we created another model that focuses only on the city of Jerusalem and manages to predict activities in Jerusalem with an accuracy level of 86% and 83% on the test set. We believe that our models can provide a good infrastructure for an app designed for tourists who come to the country and want to know what they should do during the visit.