自组织神经影射网络排序及其在植物群落分析中的应用  

Ordination of self-organizing feature map neural network and its application in the study of plant communities

作　　者：张金屯[1] 孟东平[2] 席跃翔[2] 

机构地区：[1]北京师范大学生命科学学院,北京新外大街100875 [2]山西大学黄土高原研究所

出　　处：《北京林业大学学报》2008年第1期1-5,共5页Journal of Beijing Forestry University

基　　金：国家自然科学基金(30070140);教育部骨干教师基金项目。

摘　　要：自组织神经影射网络(SOFM)对复杂问题和非线性问题具有较强的分析和解决功能,其特征完全适合植物群落的排序研究。该文介绍了SOFM的基本原理和排序分析过程及方法,并应用SOFM网络排序对太行山中段植物群落进行了排序分析。其计算过程在Matlab 6.5神经网络工具箱中实现。结果将68个样方排列在SOFM拓扑空间,排序轴反映了明确的生态梯度,能够反映植物群落间的生态关系,生态意义明确,符合植被实际,表明SOFM网络是有效的植物群落排序方法。在SOFM排序过程中也很容易进行聚类,有利于群落分类和排序的结合。The self-organizing feature map （SOFM） neural network is powerful in analyzing and solving complicated and non-linear matters. According to its features, SOFM is completely applicable to ordination study of plant communities. In the present work, the mathematical principles, ordination technique and procedure were introduced, and SOFM ordination was applied to the study of plant communities in the midst of Taihang Mountains,northern China. The ordination was carried out using NNTool box in the Matlab 6.5. As the results, 68 quadrats of plant communities were distributed in SOFM space. The ordination axes showed ecological gradients clearly and revealed the relationship among communities with ecological meanings. This result is consistent to the reality of vegetation in the study area and it suggests that SOFM ordination is an effective ordination technique in plant ecology. During the ordination procedure, it is easy to carry out clustering of communities, and so it is beneficial for combination of classification and ordination in vegetation study.

关 键 词：自组织神经影射网络 植被 数量方法 梯度分析 排序 太行山 

分 类 号：S711[农业科学—林学]