大数据处理技术在土壤肥力评价中的研究  

Study on big data processing techniques in evaluation of soil fertility

作　　者：杨玉琴[1] 陈桂芬[1] 郭宏亮[1] 孙雄辉[1] 蔡丽霞[1] 

机构地区：[1]吉林农业大学信息技术学院,长春市130118

出　　处：《中国农机化学报》2016年第4期233-236,共4页Journal of Chinese Agricultural Mechanization

基　　金：国家863项目(2006AA10A309);国家星火计划(2008GA661003);吉林省世行项目(2011-Z20);吉林省农委项目基于物联网的智能温室蔬菜大棚

摘　　要：采用K-means算法对土壤肥力数据进行聚类分析,然后针对传统K-means聚类算法在处理大数据量时时间复杂度高的难题,提出基于大数据处理技术的K-means算法。实验结果表明:(1)用K-means算法对2013年农安县13个乡镇的土壤养分数据的聚类结果显示,当k值设为3时聚类效果最优,且与实际情况相符。(2)基于Hadoop平台的MapReduce分布式下实现的K-means算法与传统的串行算法相比,提高运行速度并完成大数据量下的计算任务。This paper uses K-means algorithm to cluster analysis on the soil fertility data,then according to the traditional K-means clustering algorithm in processing large amounts of data problem of high time complexity,the K-means algorithm is proposed for based on big data processing technology.The experimental results show that（1）K-means algorithm with soil nutrient data in 2013Nong＇an 13 towns were clustering,the results show,when the K value is set to 3,the cluster effect is optimal,and consistent with the actual situation.（2）K-means algorithm under the MapReduce framework compared with the traditional serial K-means algorithms improves the operation speed and completes the computing task of large-scale data successfully.

关 键 词：K-MEANS算法 HADOOP平台 MAPREDUCE模型 土壤肥力 

分 类 号：TP399[自动化与计算机技术—计算机应用技术] S158.2[自动化与计算机技术—计算机科学与技术]