Tourism Research Based on Multi-Objective Optimization Algorithm and Random Forests

The purpose of this paper is to construct a set of dynamic and multi-dimensional sustainable tourism system through computer related technology to address the problems related to tourism. In the early stage of model construction, relevant data are collected and pre-processed scientifically with the help of data collection and processing technology, and presented intuitively through data visualization technology. In the research process, two core models are established: Model 1 is a dynamic multi-objective framework for sustainable tourism optimization, which adopts a weighted multi-objective dynamic optimization algorithm to accurately identify the sub-factors affecting income, environmental protection and tourists' satisfaction, construct a reasonable weighted objective function, and solve it by using the Sequential Least Squares Planning (SLSQP) algorithm, so that the optimal number of tourists, such as the annual number of 1.7 million tourists, the estimated annual income of 390 million U.S. dollars, and so on, can be obtained. Its annual revenue is 390 million dollars. Meanwhile, the expenditure plan is incorporated into the model, and the algorithm enables dynamic simulation and further optimization in subsequent years. Model 2 is a multi-objective optimization model based on the commercial district, which uses web crawler technology to capture coordinate data and combines with the SLSQP algorithm to determine the optimal point, and after optimization, the distance between the facility and the residential area is increased by 15.3%, the distance between the facility and the attraction is reduced by 12.8%, and the distance between the facilities is reduced by 10.4%. Using 20 sets of urban data, the random forest regression algorithm is applied to model 1 to generate heat map and coefficient matrix to visualize the influence of each factor on the objective function, so as to verify the applicability of the model. The dynamic, multi-perspective sustainable tourism model constructed based on the multi-objective optimization algorithm effectively balances the optimization factors in various aspects, provides a solution based on computer technology for solving the over-tourism problem in cities like Juno, and provides powerful data support and technical guarantee for governmental decision-making in optimizing the number of tourists and the layout of facilities, dynamic prediction and promotion of long-term development, and has high application potential and popularization value. It can provide powerful data support and technical guarantee for governmental decision-making in optimizing the number of tourists and the layout of facilities and promoting long-term development.