摘要
在Matlab7.0平台下,构建了基于减法聚类的模糊神经网络评价模型,依据6个评价指标对中国31个省(市、区)的农村经济发展水平进行了综合评价。结果表明,发展水平高于研究单元平均水平的有上海、北京、天津、浙江等12个省(市、区),低于平均发展水平的有19个省(市、区);上海市发展水平最高,北京市次之,贵州省最低。根据评价结果,应用FCM聚类法,将其发展水平分为5个层次:第一、二层次为以上海市为代表的东部7省(市),第三层次为以山东为代表的东部4省和中、西部3省(区),第四层次为以黑龙江为代表的中部7省和西部2省(市),第五层次为西部8省(区)。其发展水平具有明显的地域联系和分异特征,总体格局为东部高、中西部低,东部与中、西部之间差异明显,中、西部间的差异不大。
With the Fuzzy Neural Network appraisal models which is based on Matlab7,0 platform and subwaction clustering, the rural economic development levels in 31 provinces (municipal or district) of China were comprehensively estimated according to 6 evaluating indexes. The results showed that there are 12 provinces (municipal or district) such as Shanghai, Beijing, Tianjin and Zhejiang, with development levels higher than the studied unit average level, 19 provinces ( municipal or district) with development levels lower than the average level. Of which, Shanghai was fast, Beijing was second and Guizhou Province was the end. According to the above results, the development levels are divided into 5 grades by using the FCM cluster way, 7 provinces in eastern China with Shanghai City as their representative fall in the last and second grades; 4 provinces with Shandong as representative in eastern China and 3 provinces in middle and westem China fall in the third grade; 7 provinces with Heilongjiang as representative in middle China and 2 provinces in western China fall in the fourth grade; and 8 provinces in western China fall in the fifth grade. In conclusion, the rural development levels are obviously in connection with regions and have differentiation characteristics. The overall patterns are high in the east areas, low in the middle and west areas. The difference between the east and the middle-west is obvious, but small between the middle and west.
出处
《安徽农业科学》
CAS
北大核心
2007年第7期2152-2153,共2页
Journal of Anhui Agricultural Sciences
关键词
农村经济
评价与分类
模糊神经网络
模糊减法聚类
模糊C-均值聚类
Rural economy
Evaluation and classification
Fuzzy neural network
Fuzzy subwactive cluster
Fuzzy C-mean value cluster