Exploring the machine learning approach to the cultural and ecological sustainable development of Fuzhou tea picking opera

Abstract

The article introduces machine learning based on a support vector machine to illustrate the scenario of Fuzhou Tea Casting Opera, a folkloric non-heritage research and ecologically sustainable development, from the perspective of economic return prediction. Fine-grained information recognition and extraction of massive, unstructured UGC data are used for data collection preprocessing in this paper. Mostly, the relationship is established by analyzing the training samples to establish the connection between input and output data. The limited sample information is used to find the best compromise between the model’s complexity and learning ability. Then, search for the parameters that make the geometric interval the largest so as to construct the function and obtain the optimal hyperplane. In this paper, the kernel function is introduced using a ɛ -insensitive loss function. To obtain the decision function in different cases, two cases are modeled using support vector machine regression. Finally, the support vector machine model is used to predict the economic benefits of the Fuzhou tea-picking operation. The SVM prediction model’s prediction error of the economic benefits of tea-picking opera for the nine years from 2010 to 2018 is always maintained at a safe value below 0.035. The accuracy and stability of the prediction effect of the decision tree model and the gray prediction model are not as good as this paper’s model. It shows that the model presented in this paper is more suitable for predicting the benefits of the cultural and ecological sustainable development of Fuzhou tea-picking opera.

Keywords

• economic forecasting
• intangible cultural heritage
• support vector machine
• machine learning
• 68m11