Optimal Travel Route Recommendation Mechanism Based on Neural Networks and Particle Swarm Optimization for Efficient Tourism Using Tourist Vehicular Data

With the swift growth in tourism all around the world, it has become vital to introduce advancements and improvements to the services provided to the tourists, in order to ensure their ease of travel and satisfaction. Optimal travel route identification and recommendation is one of these amenities, which requires our attention as a basic and much-needed facility to improve the experience of travelers. In this work, we propose an optimal route recommendation mechanism for the prediction of the next tourist attraction and optimal route recommendation to the predicted tourist attraction. The algorithms used in the proposed methodology are neural networks for prediction and particle swarm optimization for finding the optimal route. We design an objective function for the route optimization based on the five route parameters of distance, road congestion, weather conditions, route popularity, and user preference. The data used is the tourism data of Jeju Island from December 2016 to December 2017. The performance analysis in the prediction mechanism is performed based on the accuracy of test data results with varying route sizes, while for route optimization, the obtained results are compared with the non-optimized technique. Also, comparisons analysis is performed by comparing the performance of the applied particle swarm optimization algorithm with an identical system-level implementation of the genetic algorithm, which is one of most widely used optimization algorithms. An extended comparative analysis with some related recommendation system studies is also performed based on key optimization factors in route optimization.