Self-organizing Map Classification Research Articles

Functional Motions Modulating VanA Ligand Binding Unraveled by Self-Organizing Maps

The VanA D-Ala:D-Lac ligase is a key enzyme in the emergence of high level resistance to vancomycin in Enterococcus species and methicillin-resistant Staphylococcus aureus. It catalyzes the formation of D-Ala-D-Lac instead of the vancomycin target, D-Ala-D-Ala, leading to the production of modified, low vancomycin binding affinity peptidoglycan precursors. Therefore, VanA appears as an attractive target for the design of new antibacterials to overcome resistance. The catalytic site of VanA is delimited by three domains and closed by an ω-loop upon enzymatic reaction. The aim of the present work was (i) to investigate the conformational transition of VanA associated with the opening of its ω-loop and of a part of its central domain and (ii) to relate this transition with the substrate or product binding propensities. Molecular dynamics trajectories of the VanA ligase of Enterococcus faecium with or without a disulfide bridge distant from the catalytic site revealed differences in the catalytic site conformations with a slight opening. Conformations were clustered with an original machine learning method, based on self-organizing maps (SOM), which revealed four distinct conformational basins. Several ligands related to substrates, intermediates, or products were docked to SOM representative conformations with the DOCK 6.5 program. Classification of ligand docking poses, also performed with SOM, clearly distinguished ligand functional classes: substrates, reaction intermediates, and product. This result illustrates the acuity of the SOM classification and supports the quality of the DOCK program poses. The protein-ligand interaction features for the different classes of poses will guide the search and design of novel inhibitors.

Image Inpainting based on Self-organizing Maps by Using Multi-agent Implementation

The image inpainting is a well-known task of visual editing. However, the efficiency strongly depends on sizes and textural neighborhood of “missing” area. Various methods of image inpainting exist, among which the Kohonen Self-Organizing Map (SOM) network as a mean of unsupervised learning is widely used. The weaknesses of the Kohonen SOM network such as the necessity for tuning of algorithm parameters and the low computational speed caused the application of multi- agent system with a multi-mapping possibility and a parallel processing by the identical agents. During experiments, it was shown that the preliminary image segmentation and the creation of the SOMs for each type of homogeneous textures provide better results in comparison with the classical SOM application. Also the optimal number of inpainting agents was determined. The quality of inpainting was estimated by several metrics, and good results were obtained in complex images.

Using self-organizing maps with complex network topologies and coalitions for time series prediction

A Self-organizing Map (SOM) is a competitive learning neural network architecture that make available a certain amount of classificatory neurons, which self-organize spatially based on input patterns. In this paper we explore the use of complex network topologies, like small-world, scale-free or random networks; for connecting the neurons within a SOM, and apply them for Time Series Prediction. We follow the VQTAM model for function prediction, and consider several benchmarks to evaluate the quality of the predictions. Afterwards, we introduce the CASOM algorithm (Coalitions and SOM) that uses coalitions to temporarily extend the neighborhood of a neuron, and to provide more accuracy in prediction problems than classical SOM. The results presented in this work suggest that the most regular the network topology is, the better results it provides in prediction. Besides, we have found that not updating all the neurons at the same time provides much better results. Finally, we describe how the use of coalitions can enhance the capability of SOM for Time Series Prediction.

Artificial neural networks assessing adolescent idiopathic scoliosis: comparison with Lenke classification

Background contextVariability in classifying and selecting levels of fusion in adolescent idiopathic scoliosis (AIS) has been repeatedly documented. Several computer algorithms have been used to classify AIS based on the geometrical features, but none have attempted to analyze its treatment patterns. PurposeTo use self-organizing maps (SOM), a kind of artificial neural networks, to reliably classify AIS cases from a large database. To analyze surgeon's treatment pattern in selecting curve regions to fuse in AIS using Lenke classification and SOM. Study designThis is a technical concept article on the possibility and benefits of using neural networks to classify AIS and a retrospective analysis of AIS curve regions selected for fusion. Patient sampleA total of 1,776 patients surgically treated for AIS were prospectively enrolled in a multicentric database. Cobb angles were measured on AIS patient spine radiographies, and patients were classified according to Lenke classification. Outcome measuresFor each patient in the database, surgical approach and levels of fusion selected by the treating surgeon were recorded. MethodsA Kohonen SOM was generated using 1,776 surgically treated AIS cases. The quality of the SOM was tested using topological error. Percentages of prediction of fusion based on Lenke classification for each patient in the database and for each node in the SOM were calculated. Lenke curve types, treatment pattern, and kappa statistics for agreement between fusion realized and fusion recommended by Lenke classification were plotted on each node of the map. ResultsThe topographic error for the SOM generated was 0.02, which demonstrates high accuracy. The SOM differentiates clear clusters of curve type nodes on the map. The SOM also shows epicenters for main thoracic, double thoracic, and thoracolumbar/lumbar curve types and transition zones between clusters. When cases are taken individually, Lenke classification predicted curve regions fused by the surgeon in 46% of cases. When those cases are reorganized by the SOM into nodes, Lenke classification predicted the curve regions to fuse in 82% of the nodes. Agreement with Lenke classification principles was high in epicenters for curve types 1, 2, and 5, moderate in cluster for curve types 3, 4, and 6, and low in transition zones between curve types. ConclusionsAn AIS SOM with high accuracy was successfully generated. Lenke classification principles are followed in 46% of the cases but in 82% of the nodes on the SOM. The SOM highlights the tendency of surgeons to follow Lenke classification principles for similar curves on the SOM. Self-organizing map classification of AIS could be valuable to surgeons because it bypasses the limitations imposed by rigid classification such as cutoff values on Cobb angle to define curve types. It can extract similar cases from large databases to analyze and guide treatment.

Soil contamination interpretation using self-organizing maps

The present study deals with the problem of soil contamination risk assessment for a region being actively impact by the routine production of Kremikovtzi metallurgical plant near Sofia, Bulgaria. Although the production is now cancelled, the soil pollution is present and needs careful assessment. The application of self-organizing maps classification strategy of Kohonen makes it possible to identify: a/ pollution sources in the region of interest; b/ spatial patterns of similarity of polluted sites and the reason for the specific pollution.

Playing with complexity: From cellular evolutionary algorithms with coalitions to self-organizing maps

Since its origins, Cellular Automata (CA) has been used to model many type of physical and computational phenomena. Interacting CAs in spatial lattices combined with evolutionary game theory have been very popular for modeling genetics or behavior in biological systems. Cellular Evolutionary Algorithms (cEAs) are a kind of evolutionary algorithm (EA) with decentralized population in which interactions among individuals are restricted to the closest ones. The use of decentralized populations in EAs allows to keep the population diversity for longer, usually resulting in a better exploration of the search space and, therefore in a better performance of the algorithm. A new adaptive technique (EACO) based on Cellular Automata, Game Theory and Coalitions uses dynamic neighborhoods to enhance the quality of cEAs. In this article we compare the characteristics EACO with classical Self-organizing Maps (SOM), and we discuss the possibilities for using Game Theory and Coalitions in the SOM scenario.

Initialization Issues in Self-organizing Maps

In this paper we present analysis and solutions to problems related to initial positioning of neurons in a classic self-organizing map (SOM) neural network. This means that we are not concerned with the multitude of growing variants, where new neurons are placed where needed. For our work, we consider placing the neurons on a Hilbert curve, as SOM have the tendency to converge similarly to self-similar curves. Another point of adjustment in SOM is the initial number of neurons, which depends on the data set. Our investigations show that initializing the neurons on a self-similar curve such as Hilbert provides a quality coverage of the input topology in much less number of epochs as compared to the usual random neuron placement. The meaning of quality is measured by absence of tangles in the network, which is one-dimensional SOM utilizing the traditional Kohonen training algorithm. The tangling of SOM presents the problem of topologically close neighbors that are actually far apart in the neuron chain of the 1D network. This is related to issues of proper clustering and analysis of cluster labels and classification. We also experiment and provide analysis where the number of neurons is concerned.

Spiking Self-organizing Maps for Classification Problem

Abstract In Self-Organizing Maps (SOM) learning, preserving the map topology to simulate the real input features is one of the most important processes. This is done by training the weight values within the Best Matching Unit (BMU) neighborhood. Improper input feeding will cause failure in identifying the potential BMU which will lead to poor map topology. Many studies have been done to optimize the structure of SOM's topology using Artificial Neural Networks (ANN).Spiking Neural Network (SNN) is the third generation of ANN, where information are transferred from one neuron to other using spikes, and processed to trigger response as an output. Current researches have proven that SNN would be an alternative solution for enhancing ANN learning due to its superiority in capturing the internal relationship of neurons. This paper proposes embedded spiking neurons for Kohonen's Self-organizing Maps (SOM) learning to improve its learning process. The proposed Spiking SOM is divided into four main phases. Phase 1 involves the development of the training sample for SOM learning through neural coding schemes. In Phase 2, the spike values are fed into the training process and potential weights are generated. Phase 3 identifies and labels the outputs from the Spiking SOM classification based on the features and characteristics. Finally, in Phase 4, proposed Spiking SOM model is validated using classification accuracy, error quantization and statistical tests using Pearson correlation. Early experiment is conducted using the 1D coding schemes for transforming dataset into spike times with hexagonal lattice structure of SOM network. Result on cancer dataset shows that the tested model has produced feasible classification accuracy with low quantization error. It shows that the 1D coding is capable in preserving the features in the input neurons.

MIGSOM: Multilevel Interior Growing Self-Organizing Maps for High Dimensional Data Clustering

Understanding the inherent structure of high-dimensional datasets is a very challenging task. This can be tackled from visualization, summarizing or simply clustering points of view. The Self-Organizing Map (SOM) is a powerful and unsupervised neural network to resolve these kinds of problems. By preserving the data topology mapped onto a grid, SOM can facilitate visualization of data structure. However, classical SOM still suffers from the limits of its predefined structure. Growing variants of SOM can overcome this problem, since they have tried to define a network structure with no need an advance a fixed number of output units by dynamic growing architecture. In this paper we propose a new dynamic SOMs called MIGSOM: Multilevel Interior Growing SOMs for high-dimensional data clustering. MIGSOM present a different architecture than dynamic variants presented in the literature. Using an unsupervised training process MIGSOM has the capability of growing map size from the boundaries as well as the interior of the network in order to represent more faithfully the structure present in a data collection. As a result, MIGSOM can have three-dimensional (3-D) structure with different levels of oriented maps developed according to data direction. We demonstrate the potential of the MIGSOM with real-world datasets of high-dimensional properties in terms of topology preserving visualization, vectors summarizing by efficient quantization and data clustering. In addition, MIGSOM achieves better performance compared to growing grid and the classical SOM.

TurSOM: A paradigm bridging Turing's unorganized machines and self-organizing maps demonstrating dual self-organization

Self-organization is a widely used technique in unsupervised learning and data analysis, largely exemplified by k-means clustering, self-organizing maps (SOM) and adaptive resonance theory. In this paper we present a new algorithm: TurSOM, inspired by Turing's unorganized machines and Kohonen's SOM. Turing's unorganized machines are an early model of neural networks characterized by self-organizing connections, as opposed to self-organizing neurons in SOM. TurSOM introduces three new mechanisms to facilitate both neuron and connection self-organization. These mechanisms are: a connection learning rate, connection reorganization, and a neuron responsibility radius. TurSOM is implemented in a 1-dimensional network (i.e. chain of neurons) to exemplify the theoretical implications of these features. In this paper we demonstrate that TurSOM is superior to the classical SOM algorithm in several ways: (1) speed until convergence; (2) independent clusters; and (3) tangle-free networks.

Forest community analysis in the Songshan national nature reserve of China using self-organizing map

The forest communities in the Songshan National Nature Reserve (SNNR) of China were analyzed by using self-organizing map (SOM). Seven types of vegetation were defined by SOM, and represented the panorama of forest communities in the SNNR. The classification of samples and species contribution was consistent. SOM classification and the dominant species visualization in the forest community data in SNNR reflected the impact of the environmental factors, such as elevation, and slope.

A novel self-organizing map (SOM) neural network for discrete groups of data clustering

The self-organizing map (SOM) neural network, also called Kohonen neural network, is an effective tool for analysis of multidimensional data. This network can be used for cluster analysis while preserving data structure (topology) in such a way that similar inputs (data) remain close together in the output layer of the network. However, no algorithm that can automatically cluster discrete groups of data is presented, and our simulation results show that the classic SOM algorithm cannot cluster discrete data correctly. In this paper, we present a novel SOM-based algorithm that can automatically cluster discrete groups of data using an unsupervised method. This method consists of three phases: at the first phase, an algorithm called “second winner” is performed, in which neurons in the competitive layer of the network find their initial location in the network space. At the second phase, a method called “batch learning” is employed, and at the end of this phase, training of the SOM network is finished. And finally at the third phase, data clustering is completed by removing the wrong links between neurons. Three real world data sets and an example of synthetic data are utilized to illustrate the accurateness and effectiveness of the proposed approach.

Adaptive FIR Neural Model for Centroid Learning in Self-Organizing Maps

In this paper, a training method for the formation of topology preserving maps is introduced. The proposed approach presents a sequential formulation of the self-organizing map (SOM), which is based on a new model of the neuron, or processing unit. Each neuron acts as a finite impulse response (FIR) system, and the coefficients of the filters are adaptively estimated during the sequential learning process, in order to minimize a distortion measure of the map. The proposed FIR-SOM model deals with static distributions and it computes an ordered set of centroids. Additionally, the FIR-SOM estimates the learning dynamic of each prototype using an adaptive FIR model. A noteworthy result is that the optimized coefficients of the FIR processes tend to represent a moving average filter, regardless of the underlying input distribution. The convergence of the resulting model is analyzed numerically and shows good properties with respect to the classic SOM and other unsupervised neural models. Finally, the optimal FIR coefficients are shown to be useful for visualizing the cluster densities.

Automatic cluster identification for environmental applications using the self-organizing maps and a new genetic algorithm

A rapid increase of environmental data dimensionality emphasizes the importance of developing data-driven inductive approaches to geographic analysis. This article uses a loosely coupled strategy to combine the technique of self-organizing maps (SOM) with a new genetic algorithm (GA) for automatic identification of clusters in multidimensional environmental datasets. In the first stage, we employ the well-known classic SOM because it is able to handle the dimensional interactions and capture the number of clusters via visualization; and thus provide extraordinary insights into original data. In the second stage, this new GA rigorously delineates the cluster boundaries using a flexibly oriented elliptical search window. To test this approach, one synthetic and two real-world datasets are employed. The results confirm a more robust and reliable approach that provides a better understanding and interpretation of massive multivariate environmental datasets, thus maximizing our insights. Other key benefits include the fact that it provides a computationally fast and efficient environment to accurately detect clusters, and is highly flexible. In a nutshell, the article presents a computational approach to facilitate knowledge discovery of massive multivariate environmental datasets; as we are too familiar with their accelerating growth rate.

Self-organizing maps classification of epidemiological data and toenail selenium content monitored on cancer and healthy patients from Poland

This paper deals with epidemiological multivariate statistical analysis of cancer and health patients from Pomeranian and Lubuskie Voivodships, Poland. The anthropometric and epidemiologic data include 8 parameters: toenail selenium concentration, sex, age, body mass index (BMI), smoking status, taking of Se supplements, health state, and family history of cancer. The self-organizing maps (SOM) are used for simultaneous classification of parameters and patients with relation to cancer diagnosis. Three different patterns (groups) of patients with cancer diagnosis are outlined: (i) older, smoking men with low toenail selenium concentration; (ii) older smoking women with family relation to cancer and toenail selenium deficiency; (iii) middle, aged nonsmokers with high level of selenium toenail concentration. The simultaneous classification of parameters and patients makes it possible to determine discriminating parameters for each pattern and relations between parameters. The relation of each parameter to cancer disease is discussed as special attention is paid to toenail selenium deficiency. More than 80% of patients with cancer diagnosis possess toenail selenium deficiency, accompanied by old age and smoking.

Static Performance of Wireless Localization Algorithm Exploiting Self-Organizing Maps

In wireless communications, determining the physical location of nodes (localization) is very important for many network services and protocols. This paper evaluates the static performance of wireless localization algorithm exploiting Self-Organizing Maps (SOM) to deal with this issue. Our proposed algorithm utilizes only connectivity information and information from some heard anchors in the network to determine the location of nodes. By introducing an efficient two-hop utilization scheme and the multi-hop anchor update, the algorithm has maximized the correlation between neighboring nodes and the global topology in distributed implementation of SOM. From our intensive simulations on various static network deployment scenarios, the results show that the proposed scheme achieves very good localization accuracy. It also reduces the SOM learning steps to around 15 to 30 steps to overcome the huge computational problem of the classical SOM.

Chemical composition of water from roofs in Gdansk, Poland

This study deals with the assessment of roof runoff waters from the region of Gdansk collected during the winter season (2007/2008). The chemical analysis includes 16 chemical variables: major ions, PAHs and PCBs measured at 3 sampling sites for 6–14 rain events. Although the data set is of limited volume the statistical data treatment using self-organizing maps (SOM) reveals the main factors controlling roof runoff water quality even for a data set with small dimension. This effort for explanation of the identified factors by the possible emission sources of the urban environment and air-particulate formation seems to be very reliable. Additionally to the roof runoff water quality factors the rain events patterns are found: “background” group of events and groups formally named “PAHs”, “PCBs” and “air-borne particles” – dominated events. The SOM classification results give an opportunity to uncover the role of roof “impact” on the runoff waters. Rain runoff water quality is described by four latent factors and the “roof” impact is uncovered.

Fuzzy optimized self-organizing maps and their application to document clustering

In this paper, an approach using fuzzy logic techniques and self-organizing maps (SOM) is presented in order to manage conceptual aspects in document clusters and to reduce the training time. In order to measure the presence degree of a concept in a document, a concept frequency formula is introduced. This formula is based on new fuzzy formulas to calculate the polysemy degree of terms and the synonymy degree between terms. In this approach, new fuzzy improvements such as automatic choice of the topology, heuristic map initialization, a fuzzy similarity measure and a keywords extraction process are used. Some experiments have been carried out in order to compare the proposed system with classic SOM approaches by means of Reuters collection. The system performance has been measured in terms of F-measure and training time. The experimental results show that the proposed approach generates good results with less training time compared to classic SOM techniques.

A dynamic neural field mechanism for self-organization

As introduced by Amari, dynamic neural fields (DNF) are a mathematical formalism aiming to describe the spatio-temporal evolution of the electrical potential of a population of cortical neurons. Various cognitive tasks have been successfully solved using this paradigm, but nevertheless, tasks requiring learning and self-organizing abilities have rarely been addressed. Aiming to extend the applicative area of DNF, we are hereby interested in using this computational model to implement such self-organizing mechanisms. Adapting the Kohonen's classical algorithm for developing self-organizing maps (SOM), we propose a DNF-driven architecture that may deploy also a self-organizing mechanism. Benefiting from the biologically-inspired features of the DNF, the advantage of such structure is that the computation is fully-distributed among its entities. Unlike the classical SOM algorithm, that requires a centralized computation of the global maximum, our proposed architecture implements a distributed decision computation, based on the local competition mechanism deployed by neural fields. Once the architecture implemented, we investigate the capacity of different neural field equations to solve simple self-organization tasks. Our analysis concludes that the considered classical equations do not perform satisfactory. Highlighting the deficiencies of these equations that impeded them to behave as expected, we propose a new system of equations, enhancing the current models, in order to handle the observed undesired effects. In summary, the novelty of these equations consist in introducing an adaptive term that triggers the re-inhibition of a so-called “unsustainable” bump of the field's activity (one that no longer is stimulated by strong input, but only by strong lateral excitation). As a conclusion, a field driven by the new equations achieves good results in solving the considered self-organizing task. Our research opens thus the way to new approaches that aim using dynamic neural field to solve more complex cognitive tasks.

Self-organizing map approaches for the haplotype assembly problem

Haplotype assembly is to reconstruct a pair of haplotypes from SNP values observed in a set of individual DNA fragments. In this paper, we focus on studying minimum error correction (MEC) model for the haplotype assembly problem and explore self-organizing map (SOM) methods for this problem. Specifically, haplotype assembly by MEC is formulated into an integer linear programming model. Since the MEC problem is NP-hard and thus cannot be solved exactly within acceptable running time for large-scale instances, we investigate the ability of classical SOMs to solve the haplotype assembly problem with MEC model. Then, aiming to overcome the limits of classical SOMs, a novel SOM approach is proposed for the problem. Extensive computational experiments on both synthesized and real datasets show that the new SOM-based algorithm can efficiently reconstruct haplotype pairs in a very high accuracy under realistic parameter settings. Comparison with previous methods also confirms the superior performance of the new SOM approach.