Abstract : Generally K-means clustering algorithm can not distinguish the
imbalance between attributes, so it can only be an independent
investigation situation of each attribute but can not be comprehensive
analysis of the soil fertility status. To solve this problem, this paper
proposes a weighted K-means clustering algorithm to evaluate the soil
fertility in Nong’an County, Jilin. The algorithm uses AHP to get
the weight of soil nutrient attributes. Then combined with K-means
clustering algorithm. Finally through the operational efficiency and
accuracy to determine the optimal classification, that can improve the
clustering algorithm of intelligent. The algorithm and the traditional
K-means clustering algorithm are used in the comparison, tests showed
that the weighted K-means clustering algorithm has a better accuracy,
operational efficiency, significantly higher than the unweighted
clustering algorithm; Comprehensive evaluation of the changes in soil
nutrients after precision fertilization that used algorithm. The soil
fertility status has a significantly improvement after years of
continuous precision fertilizing. The results show that the improved
clustering algorithm is a good method to comprehensive evaluation of
soil fertility.
https://hal.inria.fr/hal-01220818 Contributor : Hal IfipConnect in order to contact the contributor Submitted on : Tuesday, October 27, 2015 - 8:23:34 AM Last modification on : Wednesday, January 17, 2018 - 10:45:36 AM Long-term archiving on: : Thursday, January 28, 2016 - 10:19:59 AM
Guifen Chen, Lixia Cai, Hang Chen, Liying Cao, Chunan Li. Analysis and Evaluation of Soil Fertility Status Based on Weighted K-means Clustering Algorithm. 7th International Conference on Computer and Computing Technologies in Agriculture (CCTA), Sep 2013, Beijing, China. pp.89-97, ⟨10.1007/978-3-642-54341-8_10⟩. ⟨hal-01220818⟩