摘要
针对黑龙江省垦区各农场的农业机械装备水平差异性较大及数据维数高的问题,基于谱聚类算法的聚类方法,对2013-2015年垦区东部36个农场的统计数据进行了聚类分析。结合各农场农业生产总值变化速率与平均机耕面积农机总动力变化速率之间的关系,将聚类结果定义为发达农场、中等发达农场、不发达农场3类。结果表明:聚类较为准确,符合垦区农业机械装备水平差异性较大的事实,能够反应垦区农业机械装备水平现状,可为垦区未来经济的协调发展和农业机械管理等方面提供理论依据和有效建议。
With problems about the difference in agricultural mechanization level and high dimension data of Heilongjiang , proposed the clustering method which based on spectral clustering algorithm to analyze the statistics of 36 farms from 2013-2015 in the reclamation area of the East. Combined with the relationship between the farm variation rate of agricultural GDP and average area of tractor agricultural machinery total power rate of change , the result of cluster is defined developed farm, more developed farm, underdeveloped farm. The result of clustering is accurate which can consistent with the reclamation area agricultural mechanization level difference of large fact and reflect the reclamation area agricul- ture machinery and equipment level of the status quo. It can provide a theoretical basis and effective suggestions for the future in the coordinated development of economy and the agricultural machinery management.
出处
《农机化研究》
北大核心
2017年第3期26-31,共6页
Journal of Agricultural Mechanization Research
基金
公益性行业(农业)科研专项(2015-2019)
关键词
农业机械装备水平
谱聚类算法
差异
黑龙江垦区
the level of agricultural machinery and equipment
spectral clustering
difference
Heilongjiang reclamation area