基于CART决策树回归的乡村信息化水平测度模型研究

Rural Informatization Measurement Model Based on CART Regression

摘要:

根据我国农村信息化建设现状，通过CART决策树回归法对测度指标进行重要性筛选，形成农村信息化水平测度指标模型；然后运用因子分析和聚类分析，对我国31个省份农村信息化水平进行比较研究。研究结果显示：所构建的农村信息化水平测度指标模型预测准确率高，可以导向乡村振兴目标；对指标的公因子分析符合当前我国农村信息化发展特点，其中江苏、广东、浙江的农村信息化发展水平领先，沿海地区农村信息化建设较好，各区域农村信息化发展水平呈现不平衡状态。最后提出相关启示和建议，应该以发挥长三角、珠三角等城市群的辐射带动作用，加大对偏远地区农村信息资源的投入；在信息基础设施建设和信息资源环境方面，因地制宜，为乡村信息化发展提供理论参考和借鉴。

Abstract:

According to the present situation of rural informatization construction in China, this paper selects the importance of measure index by CART decision tree regression method to form the rural informatization level measurement index model, and then uses factor analysis and cluster analysis to compare the level of rural informatization in various provinces of China. The results show that the model has high prediction accuracy and can guide the goal of rural revitalization; the common factor analysis of the index accords with the current characteristics of rural informatization development in China, among which Jiangsu, Guangdong and Zhejiang have the leading level of rural informatization development, the construction of rural informatization in coastal areas is better, and the regional development of the whole country presents an unbalanced state. At the end, some suggestions are put forward such as the radiant and leading role of the Yangtze River Delta, Pearl River Delta and other urban agglomerations should be brought into play; increase the investment in rural information resources in remote areas, and adjust measures to local conditions in the information infrastructure construction and information resources environment, so as to provide theoretical reference for the development of rural informatization.