Design of travel route recommendation system based on fast Spark artificial intelligence architecture

In order to overcome the problems of low accuracy of current system tourism route prediction and poor accuracy of route recommendation, this paper proposes a tourism route recommendation system design based on fast Spark artificial intelligence architecture. In the hardware part, the search process structure of tourist route, web page information classification and extraction module and tourist document classification process structure are designed. For the software part, firstly the problem is defined for the tourist route recommendation, secondly the Markov model is constructed, and finally the tourist route is mined. By ranking the transfer probability of tourist attractions, the tourist route can be obtained, and finally the tourist route recommendation is realised. The experimental results show that the designed system has a prediction accuracy of more than 70% and a recommendation accuracy of more than 80%, with high prediction accuracy and route recommendation accuracy.