Grid topologies for the self-organizing map
Grid topologies for the self-organizing map
复制标题
DOI：
10.1016/j.neunet.2014.05.001
复制
发表时间：
2014-08-01
期刊：
影响因子：
7.8
通讯作者：
Diaz Ramos, Antonio
中科院分区：
文献类型：
--
作者：
Lopez-Rubio, Ezequiel;Diaz Ramos, Antonio
最初的自组织特征映射(SOFM)已经在许多方面进行了扩展，以适应不同的目标和应用领域。然而，我们在文献中可以找到的地图格子的拓扑几乎总是正方形的，或者更少的是六边形的。在这篇文章中，我们研究了从网格的几何理论中衍生出来的各种网格拓扑。给出了无监督聚类、彩色图像分割和分类任务的实验结果，结果表明，在大多数情况下，拓扑之间的差异在统计学上是显著的，最优拓扑取决于手头的问题。文中还对这些结果进行了理论解释。(C)2014爱思唯尔有限公司。保留所有权利。
The original Self-Organizing Feature Map (SOFM) has been extended in many ways to suit different goals and application domains. However, the topologies of the map lattice that we can found in literature are nearly always square or, more rarely, hexagonal. In this paper we study alternative grid topologies, which are derived from the geometrical theory of tessellations. Experimental results are presented for unsupervised clustering, color image segmentation and classification tasks, which show that the differences among the topologies are statistically significant in most cases, and that the optimal topology depends on the problem at hand. A theoretical interpretation of these results is also developed. (C) 2014 Elsevier Ltd. All rights reserved.