Modified self-organising maps with a new topology and initialisation algorithm

Mapping quality of the self-organising maps (SOMs) is sensitive to the map topology and initialisation of neurons. In this article, in order to improve the convergence of the SOM, an algorithm based on split and merge of clusters to initialise neurons is introduced. The initialisation algorithm speeds up the learning process in large high-dimensional data sets. We also develop a topology based on this initialisation to optimise the vector quantisation error and topology preservation of the SOMs. Such an approach allows to find more accurate data visualisation and consequently clustering problem. The numerical results on eight small-to-large real-world data sets are reported to demonstrate the performance of the proposed algorithm in the sense of vector quantisation, topology preservation and CPU time requirement.     

Simulated annealing technique for fast learning of SOM networks

The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for multidimensional input datasets. In this paper, we present an application of the simulated annealing procedure to the SOM learning algorithm with the aim to obtain a fast learning and better performances in terms of quantization error. The proposed learning algorithm is called Fast Learning Self-Organized Map, and it does not affect the easiness of the basic learning algorithm of the standard SOM. The proposed learning algorithm also improves the quality of resulting maps by providing better clustering quality and topology preservation of input multi-dimensional data. Several experiments are used to compare the proposed approach with the original algorithm and some of its modification and speed-up techniques.     

A directed batch growing approach to enhance the topology preservation of self-organizing map

The growing self-organizing map (GSOM) possesses effective capability to generate feature maps and visualizing high-dimensional data without pre-determining their size. Most of the proposed growing SOM algorithms use an incremental learning strategy. The conventional growing approach of GSOM is based on filling all available position around the candidate neuron which can decrease the topology preservation quality of the map due to the misconfiguration and twisting of the map which could be a consequence of unexpected network growth and improper neuron addition and weight initialization. To overcome this problem, in this paper we introduce a batch learning strategy for growing self-organizing maps called DBGSOM which direct the growing process based on the accumulative error around the candidate boundary neuron. In the proposed growing approach, just one new neuron is added around each candidate boundary neuron. The DBGSOM offers suitable mechanisms to find a proper growing positions and allocating initial weight vectors for the new neurons.The potential of the DBGSOM was investigated with one synthetic dataset and six real-world benchmark datasets in terms of topology preservation and mapping quality. Experimental results showed that the proposed growing strategy provides an enhanced topology preserved map and reduces the susceptibility of twisting compared to GSOM. Furthermore, the proposed method has a better clustering ability than GSOM and SOM. According to the lower number of neurons generated by DBGSOM, it needs less time to learn the manifold of the data points compared to GSOM.     

A combined measure for quantifying and qualifying the topology preservation of growing self-organizing maps

The Self-Organizing Map (SOM) is a neural network model that performs an ordered projection of a high dimensional input space in a low-dimensional topological structure. The process in which such mapping is formed is defined by the SOM algorithm, which is a competitive, unsupervised and nonparametric method, since it does not make any assumption about the input data distribution. The feature maps provided by this algorithm have been successfully applied for vector quantization, clustering and high dimensional data visualization processes. However, the initialization of the network topology and the selection of the SOM training parameters are two difficult tasks caused by the unknown distribution of the input signals. A misconfiguration of these parameters can generate a feature map of low-quality, so it is necessary to have some measure of the degree of adaptation of the SOM network to the input data model. The topology preservation is the most common concept used to implement this measure. Several qualitative and quantitative methods have been proposed for measuring the degree of SOM topology preservation, particularly using Kohonen's model. In this work, two methods for measuring the topology preservation of the Growing Cell Structures (GCSs) model are proposed: the topographic function and the topology preserving map.     

Building Adaptive Basis Functions with a Continuous Self-OrganizingMap

This paper introduces CSOM, a continuous version of the Self-Organizing Map(SOM). The CSOM network generates maps similar to those created with theoriginal SOM algorithm but, due to the continuous nature of the mapping,CSOM outperforms the SOM on function approximation tasks. CSOM integratesself-organization and smooth prediction into a single process. This is adeparture from previous work that required two training phases, one toself-organize a map using the SOM algorithm, and another to learn a smoothapproximation of a function. System performance is illustrated with threeexamples.     

Self-organizing topological tree for online vector quantization and data clustering.

The self-organizing Maps (SOM) introduced by Kohonen implement two important operations: vector quantization (VQ) and a topology-preserving mapping. In this paper, an online self-organizing topological tree (SOTT) with faster learning is proposed. A new learning rule delivers the efficiency and topology preservation, which is superior of other structures of SOMs. The computational complexity of the proposed SOTT is O(log N) rather than O(N) as for the basic SOM. The experimental results demonstrate that the reconstruction performance of SOTT is comparable to the full-search SOM and its computation time is much shorter than the full-search SOM and other vector quantizers. In addition, SOTT delivers the hierarchical mapping of codevectors and the progressive transmission and decoding property, which are rarely supported by other vector quantizers at the same time. To circumvent the shortcomings of clustering performance of classical partition clustering algorithms, a hybrid clustering algorithm that fully exploit the online learning and multiresolution characteristics of SOTT is devised. A new linkage metric is proposed which can be updated online to accelerate the time consuming agglomerative hierarchical clustering stage. Besides the enhanced clustering performance, due to the online learning capability, the memory requirement of the proposed SOTT hybrid clustering algorithm is independent of the size of the data set, making it attractive for large database.     

Incorporating self-organizing map with text mining techniques for text hierarchy generation

Self-organizing maps (SOM) have been applied on numerous data clustering and visualization tasks and received much attention on their success. One major shortage of classical SOM learning algorithm is the necessity of predefined map topology. Furthermore, hierarchical relationships among data are also difficult to be found. Several approaches have been devised to conquer these deficiencies. In this work, we propose a novel SOM learning algorithm which incorporates several text mining techniques in expanding the map both laterally and hierarchically. On training a set of text documents, the proposed algorithm will first cluster them using classical SOM algorithm. We then identify the topics of each cluster. These topics are then used to evaluate the criteria on expanding the map. The major characteristic of the proposed approach is to combine the learning process with text mining process and makes it suitable for automatic organization of text documents. We applied the algorithm on the Reuters-21578 dataset in text clustering and categorization tasks. Our method outperforms two comparing models in hierarchy quality according to users’ evaluation. It also receives better F1-scores than two other models in text categorization task.     

Automatic cluster detection in Kohonen's SOM.

Kohonen's self-organizing map (SOM) is a popular neural network architecture for solving problems in the field of explorative data analysis, clustering, and data visualization. One of the major drawbacks of the SOM algorithm is the difficulty for nonexpert users to interpret the information contained in a trained SOM. In this paper, this problem is addressed by introducing an enhanced version of the Clusot algorithm. This algorithm consists of two main steps: 1) the computation of the Clusot surface utilizing the information contained in a trained SOM and 2) the automatic detection of clusters in this surface. In the Clusot surface, clusters present in the underlying SOM are indicated by the local maxima of the surface. For SOMs with 2-D topology, the Clusot surface can, therefore, be considered as a convenient visualization technique. Yet, the presented approach is not restricted to a certain type of 2-D SOM topology and it is also applicable for SOMs having an n-dimensional grid topology.     

Clustering of the self-organizing map using a clustering validity index based on inter-cluster and intra-cluster density

The self-organizing map (SOM) has been widely used in many industrial applications. Classical clustering methods based on the SOM often fail to deliver satisfactory results, specially when clusters have arbitrary shapes. In this paper, through some preprocessing techniques for filtering out noises and outliers, we propose a new two-level SOM-based clustering algorithm using a clustering validity index based on inter-cluster and intra-cluster density. Experimental results on synthetic and real data sets demonstrate that the proposed clustering algorithm is able to cluster data better than the classical clustering algorithms based on the SOM, and find an optimal number of clusters.     

PolSOM: A new method for multidimensional data visualization

In this paper, a new algorithm named polar self-organizing map (PolSOM) is proposed. PolSOM is constructed on a 2-D polar map with two variables, radius and angle, which represent data weight and feature, respectively. Compared with the traditional algorithms projecting data on a Cartesian map by using the Euclidian distance as the only variable, PolSOM not only preserves the data topology and the inter-neuron distance, it also visualizes the differences among clusters in terms of weight and feature. In PolSOM, the visualization map is divided into tori and circular sectors by radial and angular coordinates, and neurons are set on the boundary intersections of circular sectors and tori as benchmarks to attract the data with the similar attributes. Every datum is projected on the map with the polar coordinates which are trained towards the winning neuron. As a result, similar data group together, and data characteristics are reflected by their positions on the map. The simulations and comparisons with Sammon's mapping, SOM and ViSOM are provided based on four data sets. The results demonstrate the effectiveness of the PolSOM algorithm for multidimensional data visualization.     

Local matrix adaptation in topographic neural maps

The self-organizing map (SOM) and neural gas (NG) and generalizations thereof such as the generative topographic map constitute popular algorithms to represent data by means of prototypes arranged on a (hopefully) topology representing map. Most standard methods rely on the Euclidean metric, hence the resulting clusters tend to have isotropic form and they cannot account for local distortions or correlations of data. For this reason, several proposals exist in the literature which extend prototype-based clustering towards more general models which, for example, incorporate local principal directions into the winner computation. This allows to represent data faithfully using less prototypes. In this contribution, we establish a link of models which rely on local principal components (PCA), matrix learning, and a formal cost function of NG and SOM which allows to show convergence of the algorithm. For this purpose, we consider an extension of prototype-based clustering algorithms such as NG and SOM towards a more general metric which is given by a full adaptive matrix such that ellipsoidal clusters are accounted for. The approach is derived from a natural extension of the standard cost functions of NG and SOM (in the form of Heskes). We obtain batch optimization learning rules for prototype and matrix adaptation based on these generalized cost functions and we show convergence of the algorithm. The batch optimization schemes can be interpreted as local principal component analysis (PCA) and the local eigenvectors correspond to the main axes of the ellipsoidal clusters. Thus, this approach provides a cost function associated to proposals in the literature which combine SOM or NG with local PCA models. We demonstrate the behavior of matrix NG and SOM in several benchmark examples and in an application to image compression.     

Exploiting Data Topology in Visualization and Clustering of Self-Organizing Maps

The self-organizing map (SOM) is a powerful method for visualization, cluster extraction, and data mining. It has been used successfully for data of high dimensionality and complexity where traditional methods may often be insufficient. In order to analyze data structure and capture cluster boundaries from the SOM, one common approach is to represent the SOM's knowledge by visualization methods. Different aspects of the information learned by the SOM are presented by existing methods, but data topology, which is present in the SOM's knowledge, is greatly underutilized. We show in this paper that data topology can be integrated into the visualization of the SOM and thereby provide a more elaborate view of the cluster structure than existing schemes. We achieve this by introducing a weighted Delaunay triangulation (a connectivity matrix) and draping it over the SOM. This new visualization, CONNvis, also shows both forward and backward topology violations along with the severity of forward ones, which indicate the quality of the SOM learning and the data complexity. CONNvis greatly assists in detailed identification of cluster boundaries. We demonstrate the capabilities on synthetic data sets and on a real 8D remote sensing spectral image.     

Cooperative and penalized competitive learning with application to kernel-based clustering

Competitive learning approaches with individual penalization or cooperation mechanisms have the attractive ability of automatic cluster number selection in unsupervised data clustering. In this paper, we further study these two mechanisms and propose a novel learning algorithm called Cooperative and Penalized Competitive Learning (CPCL), which implements the cooperation and penalization mechanisms simultaneously in a single competitive learning process. The integration of these two different kinds of competition mechanisms enables the CPCL to locate the cluster centers more quickly and be insensitive to the number of seed points and their initial positions. Additionally, to handle nonlinearly separable clusters, we further introduce the proposed competition mechanism into kernel clustering framework. Correspondingly, a new kernel-based competitive learning algorithm which can conduct nonlinear partition without knowing the true cluster number is presented. The promising experimental results on real data sets demonstrate the superiority of the proposed methods.     

Self-Organizing-Map Based Clustering Using a Local Clustering Validity Index

Classical clustering methods, such as partitioning and hierarchical clustering algorithms, often fail to deliver satisfactory results, given clusters of arbitrary shapes. Motivated by a clustering validity index based on inter-cluster and intra-cluster density, we propose that the clustering validity index be used not only globally to find optimal partitions of input data, but also locally to determine which two neighboring clusters are to be merged in a hierarchical clustering of Self-Organizing Map (SOM). A new two-level SOM-based clustering algorithm using the clustering validity index is also proposed. Experimental results on synthetic and real data sets demonstrate that the proposed clustering algorithm is able to cluster data in a better way than classical clustering algorithms on an SOM.     

SOMwise regression: a new clusterwise regression method

We present a novel neural learning architecture for regression data analysis. It combines, at the high level, a self-organizing map (SOM) structure, and, at the low level, a multilayer perceptron at each unit of the SOM structure. The goal is to build a clusterwise regression model, that is, a model recognizing several clusters in the data, where the dependence between predictors and response is variable (typically within some parametric range) from cluster to cluster. The proposed algorithm, called SOMwise Regression, follows closely in the spirit of the standard SOM learning algorithm and has performed satisfactorily on various test problems.     

A SOM prototype-based cluster analysis methodology

Data clustering is aimed at finding groups of data that share common hidden properties. These kinds of techniques are especially critical at early stages of data analysis where no information about the dataset is available. One of the mayor shortcomings of the clustering algorithms is the difficulty for non-experts users to configure them and, in some cases, interpret the results. In this work a computational approach with a two-layer structure based on Self-Organizing Map (SOM) is presented for cluster analysis. In the first level, a quantization of the data samples using topology-preserving metrics to automatically determine the number of units in the SOM is proposed. In the second level the obtained SOM prototypes are clustered by means of a connectivity analysis to explore the quality of the partitioning with different number of clusters. The most important benefit of this two-layer procedure is that computational load decreases considerably in comparison with data based clustering methods, making it possible to cluster large data sets and to consider several different clustering alternatives in a limited time. This methodology produces a two-dimensional map representation of the, usually, high dimensional input space, along with quantitative information on viable clustering alternatives, which facilitates the exploration of the possible partitions in a dataset. The efficiency and interpretation of the methodology is illustrated by its application to artificial, benchmark and real complex biological datasets. The experimental results demonstrate the ability of the method to identify possible segmentations in a dataset, compared to algorithms that only yield a single clustering solution. The proposed algorithm tackles the intrinsic limitations of SOM and the parameter settings associated with the clustering methodology, without requiring the number of clusters or the SOM architecture as a prerequisite, among others. This way, it makes possible its application even by researchers with a limited expertise in machine learning.     

WeVoS-ViSOM: An ensemble summarization algorithm for enhanced data visualization

This study presents a novel version of the Visualization Induced Self-Organizing Map based on the application of a new fusion algorithm for summarizing the results of an ensemble of topology-preserving mapping models. The algorithm is referred to as Weighted Voting Superposition (WeVoS). Its main feature is the preservation of the topology of the map, in order to obtain the most accurate possible visualization of the data sets under study. To do so, a weighted voting process between the units of the maps in the ensemble takes place, in order to determine the characteristics of the units of the resulting map. Several different quality measures are applied to this novel neural architecture known as WeVoS-ViSOM and the results are analyzed, so as to present a thorough study of its capabilities. To complete the study, it has also been compared with the well-know SOM and its fusion version, with the WeVoS-SOM and with two other previously devised fusion Fusion by Euclidean Distance and Fusion by Voronoi Polygon Similarity—based on the analysis of the same quality measures in order to present a complete analysis of its capabilities. All three summarization methods were applied to three widely used data sets from the UCI Repository. A rigorous performance analysis clearly demonstrates that the novel fusion algorithm outperforms the other single and summarization methods in terms of data sets visualization.     

基于寻优算法的改进自组织映射研究

为加快自组织映射的学习速度,提出一个改进的自组织映射(SOM)算法。该算法将类似模拟退火过程应用于SOM学习算法中,动态调整学习参数来优化神经元的运动,并且在损耗值达到一定阈值的情况下提前停止自组织映射聚类,保证输入数据与映射规则的快速学习与较好性能。在提高学习速度的前提下,达到输入到输出的图形一致性。在不同大容量数据集的测试结果表明,该算法与原始SOM算法及其改进算法相比,在收敛速度上可以提高一倍左右,精度上较标准SOM提高50%左右。     

基于泛化竞争和局部渗透机制的自组织网TSP问题求解方法

旅行商问题(TSP)是组合优化中最典型的NP完全问题之一,具有很强的工程背景和应用价值.文章在分析了标准SOM(Self-Organizing Map)算法在求解TSP问题的不足和在寻求总体最优解的潜力的基础上,引入泛化竞争和局部渗透这两个新的学习机制,提出了一种新的SOM算法---渗透的SOM(Infiltrative SOM,ISOM)算法.通过泛化竞争和局部渗透策略的协同作用:总体竞争和局部渗透并举、先倾向总体竞争后倾向局部渗透、在总体竞争基础上的局部渗透,实现了在总体路径寻优指导下的局部路径优化,从而使所得路径尽可能接近最优解.通过对TSPLIB中14组TSP实例的测试结果及与KNIES、SETSP、Budinich和ESOM等类SOM算法的比较,表明该算法既简单又能使解的质量得到很大提高,同时还保持了解的良好的稳健特性.     

Unsupervised learning of prototypes and attribute weights

In this paper, we introduce new algorithms that perform clustering and feature weighting simultaneously and in an unsupervised manner. The proposed algorithms are computationally and implementationally simple, and learn a different set of feature weights for each identified cluster. The cluster dependent feature weights offer two advantages. First, they guide the clustering process to partition the data set into more meaningful clusters. Second, they can be used in the subsequent steps of a learning system to improve its learning behavior. An extension of the algorithm to deal with an unknown number of clusters is also proposed. The extension is based on competitive agglomeration, whereby the number of clusters is over-specified, and adjacent clusters are allowed to compete for data points in a manner that causes clusters which lose in the competition to gradually become depleted and vanish. We illustrate the performance of the proposed approach by using it to segment color images, and to build a nearest prototype classifier.