No abstract

with the continuous improvement of people's living standard, tourism has become a popular "product". The Xiaonanhai national geological park, located in Qianjiang, Chongqing, is praised for its beautiful landscape and "the only ancient quake lake in China that is the most intact in the world". After a lot of capital investment and personnel management, the park presents good momentum, but there is still a gap between park's development and tourist demand. Finding the correct position and existing problems for its subsequent development becomes the key point of development of the scenic spot. This paper, based on market orientation theory, using the SWOT analysis method to fully analyzes the internal resources and external conditions of Xiaonanhai geological park. This paper also collects the tourist's spatial source, key visit time as well as major tourist purpose, thus countermeasures and suggestions are put forward according to these data and factors. Based on the above analysis, we put forward the corresponding improvement measures using the AISAS consumer behavior analysis model in order to optimize the Xiaonanhai national geological park market construction and to optimize the management.

The deterministic economic system will also produce chaotic dynamic behaviour, so economic chaos is getting more and more attention, and the research of economic chaos forecasting methods has become an important topic at present. The traditional economic chaos forecasting models are mostly based on large samples, but in actual production activities, there are a large number of small-sample economic chaos problems, and there is still no effective solution. This paper proposes a combined forecasting model based on the traditional economic chaos forecasting method. First of all, through the decision tree classification, priority selection of features, rough prediction is achieved. Secondly, we use BP neural network to make secondary prediction. Because the initial weight is randomly selected, it is easy to fall into the defect of local minimum. This paper optimizes the BP neural network. Finally, the decision tree model and the BP neural network model optimized by the improved genetic algorithm are combined, and the combined model is optimized by the improved GA. This method can take advantage of many prediction models and combine the prediction information of multiple different prediction models to effectively improve the fitting ability of the model and improve the prediction accuracy.