Design of Path Planning Algorithm for Tourist Attractions Based on Deep Learning

There is no reasonable tourism planning route, the number of individual tourist attractions is too large, resulting in a low tourist safety factor, and the number of tourists in a few scenic spots is scarce, which has become the biggest problem that plagues tourism companies and tourists. DL (Deep learning) based on ANN (artificial neural network), a DL structure is constructed by using multi-layer perceptron with multiple hidden layers. In this paper, the path planning algorithm of tourist attractions based on DL is designed. Aiming at the defects of BPNN (BP neural network) such as slow convergence speed and easy to fall into local minimum points, the ACO (ant colony optimization) with good global search ability is proposed. The global update formula is further optimized in the strategy based on the combination of global and local update pheromones in the improved scheme, and this global and local update pheromone scheme is applied to BPNN. The research results show that the algorithm in this paper is more effective and the optimization efficiency is higher, and the ACO-optimized BPNN not only greatly saves the complexity of operation but also improves the efficiency, and efficiently obtains the specific route of each path, providing convenient services for travel enthusiasts.