多种分类器在农用地分等中的应用及其用法改良

摘要/Abstract

摘要： 以广东省第二次土壤普查成果资料为主要数据源，选取贝叶斯决策、BP神经网络、概率神经网络、聚类等分类方法分别对数据源进行分类；并且，笔者为了充分利用有监督学习分类准确率高和无监督学习无需标定的学习样本的优点，提出了基于监督--非监督的聚类算法，然后对上述五种方法的评价结果作了比较分析；实验表明文章提出的基于监督--非监督聚类方法只利用少量的有标定学习样本，即可得到较高的分类准确率，特别在少量样本时，该方法能得到比贝叶斯决策方法、BP神经网络和概率神经网络等监督学习方法更好的土地评价结果；在实际应用中，可以尝试结合监督和非监督学习的方法，实现分类正确率和获取大量有类标签的样本之间的折中。

关键词: 水稻, 水稻, 花培, 寒地, 育种

Abstract: In this paper, the Byes method, Backpropagation Neural Network (BP), Probabilistic Neural Network (PNN) method and Clustering method are used to classify the data rooted from the data of the second earth investigation result in Guangdong province. Moreover, based-on supervised-unsupervised cluster algorithm is presented in order to take full advantage of high accuracy of supervised learning classification and no necessity demarcated study samples. Then, the estimation result of the above five methods is compared and analyzed. According to the experiment, the gaining of higher classification accuracy just needs a few of demarcated study samples if based-on supervised-unsupervised learning is used. And especially, better earth estimation result than the Byes method, BP Network, PNN method and so on, can be gotten when a few of samples are used. In land evaluation, supervised and unsupervised learning method can be combined to implement a tradeoff between classification accuracy and expense.

中图分类号:

TP18

陈其昌,薛月菊,胡月明,杨敬锋,陈志民. 多种分类器在农用地分等中的应用及其用法改良[J]. 中国农学通报, 2007, 23(2): 398-398.

Chen Qichang, Xue Yueju, Hu Yueming, Yang Jingfeng, Chen Zhimin. Multi-Classifier Application and Improved Method Research in the Land Evaluation[J]. Chinese Agricultural Science Bulletin, 2007, 23(2): 398-398.

参考文献

[1] 高中贵,彭补拙.我国农用地分等定级研究综述[J].经济地理,2004,24(4):514-519.
[2] Bhattacharya,B., Solomatine,D.P.. Machine learning in soil classification[J]. Neural Networks, 2006,19(2):186-195.
[3] Yoon Geun-Won, Park Jeong-Ho, Choi Kyoung-Ho. Land-cover supervised classification using user-oriented feature database[J], Geoscience and Remote Sensing Symposium, 2004 IEEE International[C], 2004,4(s):2724-2726.
[4] 胡月明,薛月菊,李波,等.从神经网络中抽取土地评价模糊规则[J].农业工程学报,2005,21(12):93-97.
[5] 林小莹,王占岐,殳浩,等.回归分析在农用土地分等中的应用[J].资源环境与工程,2005,19(1):54-59.
[6] 汪善勤,周勇,张甘霖.基于GIS的中国土壤分类专家系统设计[J].土壤学报,2005,42(5):705-711.
[7] 薛月菊,胡月明,等.基于概率神经网络的广东省土地资源评价.华南农业大学学报,2006,27(3):108-110.
[8] David H.Wolpert, editor. The relationship between PAC and the statistical physics.framework,the Bayesian framework and the VC framework. The mathematics of Generalization 117—214, Addision-Wesley, Reading,MA,1995.
[9] 殷柏涛.一种改进的无监督竞争学习算法.声学与电子工程,1995,(2):36-40.
[10] Q. Liu, S. E. Levinson, Y. Wu, T. S. Huang, Interactive and Incremental Learning via a Mixture of Supervised and Unsupervised Learning Strategies, Proceedings of the Fifth Joint Conference on Information Sciences, vol. 1, pp. 555558. Atlantic City, NJ, U.S.A., February 27-March 3, 2000: 1123.
[11] Richard O.Duda Peter E.Hart David G.Stork著, 李宏东,姚天翔,等译.模式分类.北京:机械工业出版社,2005:16-18.
[12] Martin T.Hagan,Howard B.Demuh,Mark H.Beale著,戴葵,等译.神经网络设计.北京:机械工业出版社,2002:197-244.
[13] Neural Network Toolbox For Use with MATLAB?, Radial Basis Networks:7-12~7-14.
[14] Jiawei Han, Micheline Kamber 著, 范明,孟小峰,等译.数据挖掘概念与技术.2004:223-227.
[15] Nong Ye, Xiang Li. A scalable, incremental learning algorithm for classification problems,Computers & Industrial Engineering 2002,(43):667-692.
[16] 司文武,钱涛.一种基于谱聚类的半监督聚类方法.计算机应用,2005,25(6):1347-1349.
[17] 胡月明.土壤资源信息系统的建立与应用[A].北京:中国科学技术出版社,2002:10.
[18] Jean-Francois Hebert,, Marc Parizeau, Nadia Ghazzali. An Hybrid Architecture for Active and Incremental Learning: The Self-Organizing Perceptron(SOP)Netwerk, IEEE, 1999:1646-1651.
[19] Xiangyang Li, Nong Ye. A Supervised Clustering and Classification Algorithm for Mining Data With Mixed Variables, IEEE Transaction on systems, March, 2006:396-406.