基于聚类分析法的广东省区域科技竞争力的评价
Evaluation of Regional Science and Technology Competitiveness of Guangdong Province Based on Cluster Analysis Method

结合广东省科技竞争力发展水平，从R&D经费、发明专利授权量、财政科技支出占地方财政支出比重等9个维度出发，构建广东省区域科技竞争力评价指标体系。综合广东统计年鉴和广东科技年鉴的数据，用K-means++方法将科技发展水平相似的城市聚为同类，并对不同类别的区域科技竞争力进行主观评价与比较。对广东省21个城市这一数据集进行多次不同训练集和测试集的选取，进行多次对比实验。实验结果表明广东省科技发展水平总体较高，但区域之间科技竞争力差异较大，科技发展水平不平衡。最后通过对各指标结果综合评价并提出了针对加强广东省科技竞争力的对策建议。
Combined with the development level of Guangdong Province’s science and technology competitiveness, this paper constructs the evaluation index system of Guangdong Province’s regional science and technology competitiveness from the following nine dimensions: R&D expenditure, invention patent authorization, and the proportion of financial science and technology expenditure in local financial expenditure etc. K-means++ method was utilized to cluster the cities in Guangdong Province through the data collected from Guangdong statistical yearbook and Guangdong science and technology yearbook. K-means++ clustered the cities with similar level of scientific and technological development into the same category, and make subjective evaluation and comparison of regional science and technology competitiveness of different categories. The dataset of 21 cities in Guangdong Province was divided into train dataset and test dataset several times. The experimental results show that the overall level of science and technology development in Guangdong province is high, but the differences in science and technology competitiveness between regions are large, and the level of science and technology development is unbalanced. Finally, through the comprehensive evaluation of the results of each index, the countermeasures and suggestions for strengthening the science and technology competitiveness of Guangdong province are put forward.