Kohonen's Self-organizing Map Algorithm Research Articles

Implementasi Jaringan Saraf Tiruan Pengenalan Pola Aksara Batak Simalungun Menggunakan Kohonen Self Organizing Map

The Simalungun Batak script contained in the Batak Letter consists of several variants of forms depending on the language and region, in general there are five variants of the Batak Letter in Sumatra, namely Karo, Toba, Pakpak, Simalungun, Angkola Mandailing. This script is not widely known by the datu, namely people who are respected by the Batak community for mastering magic, fortune-telling, and calendaring so that errors still often occur in the introduction of this script, even though this script can be found in various libraries, namely the traditional books of the Batak community. By utilizing Artificial Neural Networks and using the Kohonen Self Organizing Map method which can be implemented in the Simalungun Batak Letter Recognition Application so that it can make it easier for the public to recognize the Simalungun Batak script. The results of the study that applied an artificial neural network with the kohonen self-organizing map algorithm made the recognition of the Simalungun Batak letter pattern easy for the public to recognize and could also preserve one of the cultures, namely the Simalungun Batak script and make it easier for users to implement it into the android-based application.

An Improved Pheromone-Based Kohonen Self- Organising Map in Clustering and Visualising Balanced and Imbalanced Datasets

The data distribution issue remains an unsolved clustering problem in data mining, especially in dealing with imbalanced datasets. The Kohonen Self-Organising Map (KSOM) is one of the well-known clustering algorithms that can solve various problems without a pre- defined number of clusters. However, similar to other clustering algorithms, this algorithm requires sufficient data for its unsupervised learning process. The inadequate amount of class label data in a dataset significantly affects the clustering learning process, leading to inefficient and unreliable results. Numerous research have been conducted by hybridising and optimising the KSOM algorithm with various optimisation techniques. Unfortunately, the problems are still unsolved, especially separation boundary and overlapping clusters. Therefore, this research proposed an improved pheromonebased PKSOM algorithm known as iPKSOM to solve the mentioned problem. Six different datasets, i.e. Iris, Seed, Glass, Titanic, WDBC, and Tropical Wood datasets were chosen to investigate the effectiveness of the iPKSOM algorithm. All datasets were observed and compared with the original KSOM results. This modification significantly impacted the clustering process by improving and refining the scatteredness of clustering data and reducing overlapping clusters. Therefore, this proposed algorithm can be implemented in clustering other complex datasets, such as high dimensional and streaming data.

Application of Data Mining Methods in the Design and Creation of New Products and Services

The purpose of research is to substantiate the need to use knowledge extraction methods in the design and creation of new products and services and the feasibility of using the Kohonen self-organizing map method through its formation. Such a map helps to identify previously unknown groups, in particular, as in the case of this article – consumer groups, and their analysis will make it possible to form new tariffs for the services of the mobile operator’s billing system. The main reason for the research is to show organizations the ability to design and create innovative products.Research methods are empirical in nature, based on the collection and accumulation of data on consumer behavior in the market and their subsequent analysis. In order to analyze the collected data, Data Mining methods are used, in particular, the Kohonen self-organizing map method, which allows to obtain automatic clustering of consumers in the market by various characteristics. Clustering was performed using the Kohonen self-organizing map algorithm implemented in the BaseGroup Labs Deductor Studio analytical platform. The choice of this software product is explained by a clear interface and the availability of the required functionality. The study was based on data provided by the mobile operator’s billing system. This is a fairly large amount of data showing the completed operations of mobile operator subscribers.Results. The article provides an overview of sources that offer possible methods for extracting knowledge and ways to process it. The Kohonen map is also built, which allows you to get information about the current situation for mobile subscribers from various independently selected areas. After analyzing this information, the revealed knowledge is applied in the formation of new tariffs and services of the mobile operator. This method of extracting knowledge can also be applied to other large volumes of data from various fields of activity. However, there is a limitation when using this type of knowledge extraction, which is that the data must be structured. If you use unstructured data, you can consider other methods for extracting knowledge described in this article.Conclusion. The article considers the stage of knowledge extraction when designing and creating new products and services based on Data Mining methods, in particular the self-organizing Kohonen map. Innovation in the design and creation of products and services is emphasized by the variability of data in accordance with the dynamic behavior of consumers in the market, which causes the need to periodically review the requirements and concepts of products and services brought to the market.

Identifying patterns in the multitrophic community and food-web structure of a low-turbidity temperate estuarine bay

Food web dynamics outline the ecosystem processes that regulate community structure. Challenges in the approaches used to capture topological descriptions of food webs arise due to the difficulties in collecting extensive empirical data with temporal and spatial variations in community structure and predator–prey interactions. Here, we use a Kohonen self-organizing map algorithm (as a measure of community pattern) and stable isotope-mixing models (as a measure of trophic interaction) to identify food web patterns across a low-turbidity water channel of a temperate estuarine-coastal continuum. We find a spatial difference in the patterns of community compositions between the estuarine and deep-bay channels and a seasonal difference in the plankton pattern but less in the macrobenthos and nekton communities. Dietary mixing models of co-occurring dominant taxa reveal site-specific but unchanging food web topologies and the prominent role of phytoplankton in the trophic base of pelagic and prevalent-detrital benthic pathways. Our approach provides realistic frameworks for linking key nodes from producers to predators in trophic networks.

An Electronic Registration for Undergraduate Students with Department Selection Based on Artificial Neural Network

The objective of the present research is to facilitate the administrative procedures associated with student registration process, and to ensure equal opportunities for all applicants in college. It aims to assist students in identifying appropriate alternatives available to the departments electronically anytime and anywhere saving time and effort for the student. For this purpose, an intelligent e-government system named "An Electronic Intelligent Registration with Department Selection" (E-IRDS) is designed as one of the intelligent e- services in Iraqi e-governance by using many tools and programming languages which are (PHP, MYSQL, HTML, CSS, XML, NOTPAD++, C#). Artificial neural networks (ANNs) technology is applied, notably Kohonen's self-organizing map (SOM) as one of the important unsupervised classification algorithms of machine learning for classifying and distributing the students automatically into the college academic departments based on their desires, their total degrees, and according to scientific plan for each department, in addition to the specific and personal student information. The applied results based on international standards demonstrated the accuracy of Kohonen's SOM algorithm in classification and distribution methods at least time and possible learning ratio. The system test and assessment results confirmed that it is characterized with a very high security and reliability and accuracy. It is also distinguished with very high efficiency and transparency as well as flexibility and high performance speed. The results also emphasized the ease and availability of the system to all students, besides the possibility of troubleshot and correct errors easily.

A MODIFIED KOHONEN SELF-ORGANIZING MAP (KSOM) CLUSTERING FOR FOUR CATEGORICAL DATA

The Kohonen Self-Organizing Map (KSOM) is one of the Neural Network unsupervised learning algorithms. This algorithm is used in solving problems in various areas, especially in clustering complex data sets. Despite its advantages, the KSOM algorithm has a few drawbacks; such as overlapped cluster and non-linear separable problems. Therefore, this paper proposes a modified KSOM that inspired from pheromone approach in Ant Colony Optimization. The modification is focusing on the distance calculation amongst objects. The proposed algorithm has been tested on four real categorical data that are obtained from UCI machine learning repository; Iris, Seeds, Glass and Wisconsin Breast Cancer Database. From the results, it shows that the modified KSOM has produced accurate clustering result and all clusters can clearly be identified.

Structure determinants of indolin-2-on-3-spirothiazolidinones as MptpB inhibitors: An in silico study

Tuberculosis caused by Mycobacterium tuberculosis (Mtb), is a major factor of death around the world. Mtb protein tyrosine phosphatase B (MptpB), a virulent phosphatase encoded by Mtb, is a potential target for the treatment of tuberculosis. To explore the structure features of 134 indolin-2-on-3-spirothiazolidinones as MptpB inhibitors, in this work, a set of ligand- and receptor-based 3D-QSAR studies were, for the first time, carried out using comparative molecular field analysis (CoMFA) and comparative molecular similarity index analysis (CoMSIA) techniques. Splitting of the training and test sets was constructed by Kohonen's self-organizing map algorithm. Additionally, to profile the possible binding modes of these inhibitors, molecule docking and molecular dynamics (MD) simulations were performed as well. The optimal QSAR models built on both ligand- and receptor-based alignments exhibit proper reliabilities and predictive abilities as illustrated by their statistical results that for ligand-based alignment CoMFA and CoMSIA models obtain Q2 = 0.662, Rncv2 = 0.799, Rpre2 = 0.757, and Q2 = 0.736, Rncv2 = 0.847, Rpre2 = 0.866; whereas for docking-based one achieve Q2 = 0.536, Rncv2 = 0.894, Rpre2 = 0.823, and Q2 = 0.498, Rncv2 = 0.812, Rpre2 = 0.833, respectively. Furthermore, the contours generated from these models are well consistent with the docking and MD results we performed. Our findings are: (1) Indolin-2-on-3-spirothiazolidinones may bind to the MptpB with a “cockhorse” conformation, which is stabilized by eight H-bonds formed with the surrounding residues (Arg166, Cys160, Ala162, Lys164, His94 and Glu129), one cation–π interaction with the side chain of Arg166 and two hydrophobic interactions in the binding site. (2) Introduction of bulky and electropositive substituents at R3, hydrophobic substituents around rings A and E can improve the inhibition potency. (3) Nitro group at R2 position of 2-oxindole core plays a crucial role for the biological activities, and may also be essential to the water-mediated H-bond networks. (4) For 8, 10-positions of the thiazolidinone ring and 13-position of isoxazole, the groups acting as H-bond acceptor are favorable for increasing the activity. All these results are expected to be useful in the discovery of novel and more potent MptpB inhibitors.

Recognize Handwritten Urdu Script Using Kohenen Som Algorithm

In this paper we use the Kohonen neural network based Self Organizing Map (SOM) algorithm for Urdu Character Recognition. Kohenen NN have more efficient in terms of performance as compare to other approaches. Classification is used to recognize hand written Urdu character. The number of possible unknown character is reducing by pre-classification with respect to subset of the total character set. So the proposed algorithm is attempt to group similar character. Members of pre-classified group are further analyzed using a statistical classifier for final recognition. A recognition rate of around 79.9% was achieved for the first choice and more than 98.5% for the top three choices. The result of this paper shows that the proposed Kohonen SOM algorithm yields promising output and feasible with other existing techniques.

Self-organization of the batch Kohonen network under quantization effects

In this work, we investigate the effect of quantized weights and inputs on the self-organizing properties of the batch variant of Kohonen's self-organizing map algorithm. In particular, we examine necessary and sufficient conditions that ensure self-organization of the batch SOM algorithm for one-dimensional (1D) networks mapping a quantized 1D input space. Using Markov chain formalism, it is shown that the existing analysis for the original algorithm can be extended to also include the more general batch variant. Finally, simulations verify the theoretical results, relate the speed of weight ordering to the distribution of the inputs, extend the results to the 2D case, and show the existence of metastable states of the Markov chain.

A novel hardware-oriented Kohonen SOM image compression algorithm and its FPGA implementation

Kohonen self-organizing map (K-SOM) has proved to be suitable for lossy compression of digital images. The major drawback of the software implementation of this technique is its very computational intensive task. Fortunately, the structure is fairly easy to convert into hardware processing units executing in parallel. The resulting hardware system, however, consumes much of a microchip’s internal resources, i.e. slice registers and look-up table units. This results in utilising more than a single microchip to realize the structure in pure hardware implementation. Previously proposed K-SOM realizations were mainly targetted on implementing on an application specific integrated circuit (ASIC) with low restriction on resource utilization. In this paper, we propose an alternative architecture of K-SOM suitable for moderate density FPGAs with acceptable image quality and frame rate. In addition, its hardware architecture and synthesis results are presented. The proposed K-SOM algorithm compromises between the image quality, the frame rate throughput, the FPGA’s resource utilization and, additionally, the topological relationship among neural cells within the network. The architecture has been proved to be successfully synthesized on a single moderate resource FPGA with acceptable image quality and frame rate.

Social area analysis, data mining, and GIS

There is a long cartographic tradition of describing cities through a focus on the characteristics of their residents. A review of the history of this type of urban social analysis highlights some persistent challenges. In this paper existing geodemographic approaches are extended through coupling the Kohonen Self-Organizing Map algorithm (SOM), a data-mining technique, with geographic information systems (GIS). This approach allows the construction of linked maps of social (attribute) and geographic space. This novel type of geodemographic classification allows ad hoc hierarchical groupings and exploration of the relationship between social similarity and geographic proximity. It allows one to filter complex demographic datasets and is capable of highlighting general social patterns while retaining the fundamental social fingerprints of a city. A dataset describing 79 attributes of the 2217 census tracts in New York City is analyzed to illustrate the technique. Pairs of social and geographic maps are formally compared using simple pattern metrics. Our analysis of New York City calls into question some assumptions about the functional form of spatial relationships that underlie many modeling and statistical techniques.

Relative characterization of rosemary samples according to their geographical origins using microwave-accelerated distillation, solid-phase microextraction and Kohonen self-organizing maps

For centuries, rosemary (Rosmarinus officinalis L.) has been used to prepare essential oils which, even now, are highly valued due to their various biological activities. Nevertheless, it has been noted that these activities often depend on the origin of the rosemary plant and the method of extraction. Since both of these quality parameters can greatly influence the chemical composition of rosemary oil, an original analytical method was developed where "dry distillation" was coupled to headspace solid-phase microextraction (HS-SPME) and then a data mining technique using the Kohonen self-organizing map algorithm was applied to the data obtained. This original approach uses the newly described microwave-accelerated distillation technique (MAD) and HS-SPME; neither of these techniques require external solvent and so this approach provides a novel "green" chemistry sampling method in the field of biological matrix analysis. The large data set obtained was then treated with a rarely used chemometric technique based on nonclassical statistics. Applied to 32 rosemary samples collected at the same time from 12 different sites in the north of Algeria, this method highlighted a strong correlation between the volatile chemical compositions of the samples and their origins, and it therefore allowed the samples to be grouped according to geographical distribution. Moreover, the method allowed us to identify the constituents that exerted the most influence during classification.

Self-organisation can generate the discontinuities in the somatosensory map

The primary somatosensory cortex contains a topographic map of the body surface, with two notable discontinuities—the representation of the face is next to that of the hands, and that of the feet is next to the genitals. Farah [Why does the somatosensory homunculus have hands next to face and feet next to genitals? a hypothesis. Neural Computation 10(8) (1998) 1983–1985] has suggested that these discontinuities are due to the mechanisms of self-organisation which underlie cortical map development. The typical position of the foetus in the womb means that these two pairs of body parts will often touch and hence their representations will be simultaneously co-active, even though they are distal in terms of the body surface. We use the Kohonen self-organising map algorithm to provide an existence proof of the plausibility of Farah's hypotheses. We then use the model to test the viability of other possible causes of the known map structure and to explore the limitations of self-organisation for explaining the features of the somatosensory map. The model shows that (a) the Kohonen algorithm requires high frequencies of co-activation to introduce a selective discontinuity into the map, and (b) that higher frequency of separate activation of the critical patterns alone is not sufficient to generate the selective discontinuity and (c) the consistency of near-optimal map formation, and in particular the medial–lateral ordering, cannot be reliably generated by a simple Kohonen algorithm.

The use of interactive graphical maps for browsing medical/health Internet information resources

As online information portals accumulate metadata descriptions of Web resources, it becomes necessary to develop effective ways for visualising and navigating the resultant huge metadata repositories as well as the different semantic relationships and attributes of described Web resources. Graphical maps provide a good method to visualise, understand and navigate a world that is too large and complex to be seen directly like the Web. Several examples of maps designed as a navigational aid for Web resources are presented in this review with an emphasis on maps of medical and health-related resources. The latter include HealthCyberMap maps , which can be classified as conceptual information space maps, and the very abstract and geometric Visual Net maps of PubMed (for demos). Information resources can be also organised and navigated based on their geographic attributes. Some of the maps presented in this review use a Kohonen Self-Organising Map algorithm, and only HealthCyberMap uses a Geographic Information System to classify Web resource data and render the maps. Maps based on familiar metaphors taken from users' everyday life are much easier to understand. Associative and pictorial map icons that enable instant recognition and comprehension are preferred to geometric ones and are key to successful maps for browsing medical/health Internet information resources.

Pockets Full of Memories: an interactive museum installation

Conceived as an installation on the topic of the archive and memory, Pockets Full of Memories was exhibited on the main floor of the Centre Pompidou National Museum of Modern Art, Paris from 10 April to 3 September 2001. During this time, approximately 20,000 visitors came to view the installation and contributed over 3300 objects in their possession, digitally scanning and describing them. This information was stored in a database and organized by the Kohonen Self-Organizing Map algorithm [http://www.cis.hut.fi/teuvo/] that positioned objects of similar descriptions near each other in a two-dimensional map. The map of objects was projected in the gallery space and was also accessible online at [www.pocketsfullofmemories.com] where individuals in the gallery and at home could review the objects and add comments and stories to any of them.

A parallel general implementation of Kohonen's self-organizing map algorithm: performance and scalability

Kohonen's self-organizing map algorithm provides computational neurobiology with a useful model of the primate cerebral cortex. However, simulations of only modestly sized maps quickly exceed the capacity of even very fast workstations. Here, we report that a parallel implementation of the algorithm on a Beowulf commodity-class computing cluster scales very favorably with the number of available nodes and greatly speeds the computation of medium-to-large-scale cortical maps.

Visualizing plastic change in a large model of somatosensory cortex using an adaptive coordinates algorithm

The neuronal empiricism hypothesis proposes that cortical neurons learn their receptive field properties and organization from statistical regularities in the stream of information that they receive. We computationally examine this hypothesis using Kohonen's self-organizing map algorithm with simulated input from the hand. By using a new adaptive coordinate algorithm we show that the map, once formed, remains stable despite inherent plasticity as long as the statistics of the sensory data do not change. If they do change, our method reveals the presence of powerful attractor regions in the map governing its plastic response.

Quantitative analysis of kernel properties in Kohonen's self-organizing map algorithm: Gaussian and difference of Gaussians neighborhoods

Recent experimental evidence suggests that processes of plastic reconfiguration of the primate cerebral cortex may involve an inhibitory component. Here, we modify Kohonen's self-organizing map model of the cortex to include surround inhibition in its adaptation kernel. This addition not only improves the accuracy of the cortical representation, as measured by quantization error, it also tends to produce pinwheel patterns, similar to those observed in primary visual cortex.

Verifying the Proximity and Size Hypothesis for Self-Organizing Maps

:The Kohonen Self-Organizing Map (SOM) is an unsupervised learning technique for summarizing high-dimensional data so that similar inputs are, in general, mapped close to one another. When applied to textual data, SOM has been shown to be able to group together related concepts in a data collection and to present major topics within the collection with larger regions. This article presents research in which we sought to validate these properties of SOM, called the Proximity and Size Hypotheses, through a user evaluation study. Building upon our previous research in automatic concept generation and classification, we demonstrated that the Kohonen SOM was able to perform concept clustering effectively, based on its concept precision and recall7 scores as judged by human experts. We also demonstrated a positive relationship between the size of an SOM region and the number of documents contained in the region. We believe this research has established the Kohonen SOM algorithm as an intuitively appealing and promising neural-network-based textual classification technique for addressing part of the longstanding “information overload” problem.

Adaptive learning method in self-organizing map for edge preserving vector quantization

The Kohonen's self-organizing map algorithm for vector quantization of images is modified to reduce the edge degradation in the coded image. The learning procedure is performed by adaptive learning rates that are determined according to the image block activity. The simulation result of 4x4 vector quantization for 512x512 image coding demonstrates the feasibility of the proposed method.