Research on Optimization and Improvement of Intelligent Management System based on Big Data Mining and ant Colony Algorithm

Abstract

With the development of social economy, the market mechanism tends to be perfect, which promotes the tourism industry to a new height. With the rapid development of tourism industry, management problems have become increasingly prominent. This paper makes an in-depth analysis of tourism planning and management from multiple perspectives. In view of the poor performance of the original optimal tourism route optimization model in obtaining the shortest route, this paper constructs the optimal tourism route optimization model based on ant colony optimization algorithm, sets the route selection process, and uses ant colony algorithm to complete the optimal route selection. According to the results of route selection, pheromone update rules and route model format are set to complete the construction of optimal route optimization model. Compared with the traditional model, the path chosen by the model is shorter and the cost is lower. At the same time, using BP neural network model and matlab calculation program to evaluate tourism resources can avoid the influence of subjective factors on the evaluation results to the greatest extent. On this basis, the evaluation model is designed, and the error value of the evaluation model is analyzed. This paper mainly focuses on the relevant measures of tourism management and puts forward a tourist flow forecasting model based on data mining. Firstly, the historical data of tourism flow are collected, and then the chaos algorithm is introduced to construct the learning sample of tourism flow prediction. Finally, the particle swarm optimization algorithm is introduced to optimize the parameters of the tourist flow forecasting model. The simulation results show that, compared with the BP neural network optimized by particle swarm optimization and support vector machine, this model can describe the changing characteristics of passenger flow in scenic spots more accurately, and the prediction error of passenger flow in scenic spots is much smaller than that of the contrast model, and a more ideal passenger flow prediction result is obtained, which can put forward a new solution strategy in the field of tourism management.