CLUSTERING PROPERTIES OF FUZZY KOHONEN'S SELF-ORGANIZING FEATURE MAPS

A new clustering algorithm called fuzzy self-organizing feature maps is introduced. It can process not only the exact digital inputs, but also the inexact or fuzzy non-digital inputs, such as natural language inputs. Simulation results show that the new algorithm is superior to original Kohonen’s algorithm in clustering performance and learning rate.

Weighted Particle Swarm Clustering Algorithm for Self-Organizing Maps

The traditional K-means clustering algorithm is difficult to determine the cluster number,which is sensitive to the initialization of the clustering center and easy to fall into local optimum.This paper proposes a clustering algorithm based on self-organizing mapping network and weight particle swarm optimization SOM&WPSO(Self-Organization Map and Weight Particle Swarm Optimization).Firstly,the algorithm takes the competitive learning mechanism of a self-organizing mapping network to divide the data samples into coarse clusters and obtain the clustering center.Then,the obtained clustering center is used as the initialization parameter of the weight particle swarm optimization algorithm.The particle position of the WPSO algorithm is determined by the traditional clustering center is improved to the sample weight,and the cluster center is the“food”of the particle group.Each particle moves toward the nearest cluster center.Each iteration optimizes the particle position and velocity and uses K-means and K-medoids recalculates cluster centers and cluster partitions until the end of the algorithm convergence iteration.After a lot of experimental analysis on the commonly used UCI data set,this paper not only solves the shortcomings of K-means clustering algorithm,the problem of dependence of the initial clustering center,and improves the accuracy of clustering,but also avoids falling into the local optimum.The algorithm has good global convergence.

Application of Self-Organizing Feature Map Neural Network Based on K-means Clustering in Network Intrusion Detection

Due to the widespread use of the Internet,customer information is vulnerable to computer systems attack,which brings urgent need for the intrusion detection technology.Recently,network intrusion detection has been one of the most important technologies in network security detection.The accuracy of network intrusion detection has reached higher accuracy so far.However,these methods have very low efficiency in network intrusion detection,even the most popular SOM neural network method.In this paper,an efficient and fast network intrusion detection method was proposed.Firstly,the fundamental of the two different methods are introduced respectively.Then,the selforganizing feature map neural network based on K-means clustering(KSOM)algorithms was presented to improve the efficiency of network intrusion detection.Finally,the NSLKDD is used as network intrusion data set to demonstrate that the KSOM method can significantly reduce the number of clustering iteration than SOM method without substantially affecting the clustering results and the accuracy is much higher than Kmeans method.The Experimental results show that our method can relatively improve the accuracy of network intrusion and significantly reduce the number of clustering iteration.

Analysis of morphological characteristics of gravels based on digital image processing technology and self-organizing map

A comprehensive understanding of spatial distribution and clustering patterns of gravels is of great significance for ecological restoration and monitoring.However,traditional methods for studying gravels are low-efficiency and have many errors.This study researched the spatial distribution and cluster characteristics of gravels based on digital image processing technology combined with a self-organizing map(SOM)and multivariate statistical methods in the grassland of northern Tibetan Plateau.Moreover,the correlation of morphological parameters of gravels between different cluster groups and the environmental factors affecting gravel distribution were analyzed.The results showed that the morphological characteristics of gravels in northern region(cluster C)and southern region(cluster B)of the Tibetan Plateau were similar,with a low gravel coverage,small gravel diameter,and elongated shape.These regions were mainly distributed in high mountainous areas with large topographic relief.The central region(cluster A)has high coverage of gravels with a larger diameter,mainly distributed in high-altitude plains with smaller undulation.Principal component analysis(PCA)results showed that the gravel distribution of cluster A may be mainly affected by vegetation,while those in clusters B and C could be mainly affected by topography,climate,and soil.The study confirmed that the combination of digital image processing technology and SOM could effectively analyzed the spatial distribution characteristics of gravels,providing a new mode for gravel research.

Semi-Supervised Clustering Algorithm Based on Deep Feature Mapping

Clustering analysis is one of the main concerns in data mining.A common approach to the clustering process is to bring together points that are close to each other and separate points that are away from each other.Therefore,measuring the distance between sample points is crucial to the effectiveness of clustering.Filtering features by label information and mea-suring the distance between samples by these features is a common supervised learning method to reconstruct distance metric.However,in many application scenarios,it is very expensive to obtain a large number of labeled samples.In this paper,to solve the clustering problem in the few supervised sample and high data dimensionality scenarios,a novel semi-supervised clustering algorithm is proposed by designing an improved prototype network that attempts to reconstruct the distance metric in the sample space with a small amount of pairwise supervised information,such as Must-Link and Cannot-Link,and then cluster the data in the new metric space.The core idea is to make the similar ones closer and the dissimilar ones further away through embedding mapping.Extensive experiments on both real-world and synthetic datasets show the effectiveness of this algorithm.Average clustering metrics on various datasets improved by 8%compared to the comparison algorithm.

Hydrogeochemical characterization and quality assessment of groundwater using self-organizing maps in the Hangjinqi gasfield area,Ordos Basin,NW China

Water resources are scarce in arid or semiarid areas,which not only limits economic development,but also threatens the survival of mankind.The local communities around the Hangjinqi gasfield depend on groundwater sources for water supply.A clear understanding of the groundwater hydrogeochemical characteristics and the groundwater quality and its seasonal cycle is invaluable and indispensable for groundwater protection and management.In this study,self-organizing maps were used in combination with the quantization and topographic errors and K-means clustering method to investigate groundwater chemistry datasets.The Piper and Gibbs diagrams and saturation index were systematically applied to investigate the hydrogeochemical characteristics of groundwater from both rainy and dry seasons.Further,the entropy-weighted theory was used to characterize groundwater quality and assess its seasonal variability and suitability for drinking purposes.Our hydrochemical groundwater dataset,consisting of 10 parameters measured during both dry and rainy seasons,was classified into 6 clusters,and the Piper diagram revealed three hydrochemical facies:Cl-Na type(clusters 1,2 and 3),mixed type(clusters 4 and 5),and HCO3-Ca type(cluster 6).The Gibbs diagram and saturation index suggested thatweathering of rock-forming mineralswere the primary process controlling groundwater chemical composition and validated the credibility and practicality of the clustering results.Two-thirds of 45 groundwater samples were categorized as excellent-or good-quality and were suitable as drinking water.Cluster changes within the same and different clusters from the dry season to the rainy season were detected in approximately 78%of the collected samples.The main factors affecting the groundwater quality were hydrogeochemical characteristics,and dry season groundwater quality was better than rainy season groundwater quality.Based on this work,such results can be used to investigate the seasonal variation of hydrogeochemical characteristics and assess water quality accurately in the others similar area.

Adaptive Surrogate Model Based Optimization (ASMBO) for Unknown Groundwater Contaminant Source Characterizations Using Self-Organizing Maps

Characterization of unknown groundwater contaminant sources in terms of location, magnitude and duration of source activity is a complex problem. In this study, to increase the efficiency and accuracy of source characterization an alternative methodology to the methodologies proposed earlier is developed. This methodology, Adaptive Surrogate Modeling Based Optimization (ASMBO) uses the capabilities of Self Organizing Map (SOM) algorithm to design the surrogate models and adaptive surrogate models for source characterization. The most important advantage of this methodology is its direct utilization for groundwater contaminant characterization without the necessity of utilizing a linked simulation optimization model. The validation of the SOM based surrogate models and SOM based adaptive surrogate models demonstrates that the quantity and quality of initial sample sizes have crucial role on the accuracy of solutions as the designed monitoring locations. The performance evaluation results of the proposed methodology are obtained using error free and erroneous concentration measurement data. These results demonstrate that the developed methodology could approximate groundwater flow and transport simulation models, and substitute the optimization model for characterization of unknown groundwater contaminant sources in terms of location, magnitude and duration of source activity.

Fault diagnosis of rocket engine ground testing bed with self-organizing maps(SOMs)

To solve the fault diagnosis problem of liquid propellant rocket engine ground testing bed,a fault diagnosis approach based on self-organizing map(SOM)is proposed.The SOM projects the multidimensional ground testing bed data into a two-dimensional map.Visualization of the SOM is used to cluster the ground testing bed data.The out map of the SOM is divided to several regions.Each region is represented for one fault mode.The fault mode of testing data is determined according to the region of their labels belonged to.The method is evaluated using the testing data of a liquid-propellant rocket engine ground testing bed with sixteen fault states.The results show that it is a reliable and effective method for fault diagnosis with good visualization property.

Extending self-organizing maps for supervised classification of remotely sensed data

An extended self-organizing map for supervised classification is proposed in this paper. Unlike other traditional SOMs, the model has an input layer, a Kohonen layer, and an output layer. The number of neurons in the input layer depends on the dimensionality of input patterns. The number of neurons in the output layer equals the number of the desired classes. The number of neurons in the Kohonen layer may be a few to several thousands, which depends on the complexity of classification problems and the classification precision. Each training sample is expressed by a pair of vectors ： an input vector and a class codebook vector. When a training sample is input into the model, Kohonen＇s competitive learning rule is applied to selecting the winning neuron from the Kohouen layer and the weight coefficients connecting all the neurons in the input layer with both the winning neuron and its neighbors in the Kohonen layer are modified to be closer to the input vector, and those connecting all the neurons around the winning neuron within a certain diameter in the Kohonen layer with all the neurons in the output layer are adjusted to be closer to the class codebook vector. If the number of training sam- ples is sufficiently large and the learning epochs iterate enough times, the model will be able to serve as a supervised classifier. The model has been tentatively applied to the supervised classification of multispectral remotely sensed data. The author compared the performances of the extended SOM and BPN in remotely sensed data classification. The investigation manifests that the extended SOM is feasible for supervised classification.

Precipitation Regionalization Using Self-Organizing Maps for Mumbai City, India

The detailed analysis of individual rain events characteristics is an essential step for improving our understanding of variation in precipitation over different topographies. In this study, the homogeneity among rain gauges was investigated using the concept of “rain event properties,” linking them to the main atmospheric system that affects the rainfall in the region. For this, eight properties of more than 23,000 rain events recorded at 47 meteorological stations in Mumbai, India, were analyzed utilizing seasonal (June-September) rainfall records over 2006-2016. The high similarities among the properties indicated the similarities among the rain gauges. Furthermore, similar rain gauges were distinguished, investigated and characterized by cluster analysis using self-organizing maps (SOM). The cluster analysis results show six clusters of similarly behaving rain gauges, where each cluster addresses one isolated class of variables for the rain gauge. Additionally, the clusters confirm the spatial variation of rainfall caused by the complex topography of Mumbai, comprising the flatland near the Arabian Sea, high-rise buildings (urban area) and mountain and hills areas (Sanjay Gandhi National Park located in the northern part of Mumbai).

Intraseasonal variability of the equatorial Pacific Ocean and its relationship with ENSO based on Self-Organizing Maps analysis

We investigated the intraseasonal variability of equatorial Pacific subsurface temperature and its relationship with El Nino-Southern Oscillation(ENSO) using Self-Organizing Maps(SOM) analysis.Variation in intraseasonal subsurface temperature is mainly found along the thermocline.The SOM patterns concentrate in basin-wide seesaw or sandwich structures along an east-west axis.Both the seesaw and sandwich SOM patterns oscillate with periods of 55 to 90 days,with the sequence of them showing features of equatorial intraseasonal Kelvin wave,and have marked interannual variations in their occurrence frequencies.Further examination shows that the interannual variability of the SOM patterns is closely related to ENSO;and maxima in composite interannual variability of the SOM patterns are located in the central Pacific during CP El Nino and in the eastern Pacific during EP El Nino.The se results imply that some of the ENSO forcing is manife sted through changes in the occurrence frequency of intraseasonal patterns,in which the change of the intraseasonal Kelvin wave plays an important role.

Self-Organizing Maps in Seismic Image Segmentation

Unsupervised neural networks such as the Kohonen Self-Organizing Maps （SOM） have been widely used for searching natural clusters in multidimensional and massive data. One example where the data available for analysis can be extremely large is seismic interpretation for hydrocarbon exploration. In order to assist the interpreter in identifying characteristics of interest confined in the seismic data, the authors present a set of data attributes that can be used to train a SOM in such a way that zones of interest can be automatically identified or segmented, reducing time in the interpretation process. The authors show how to associate SOM to 2D color maps to visually identify the clustering structure of the input seismic data, and apply the proposed technique to a 2D synthetic seismic dataset of salt structures.

Selecting Alternatives from Self-Organizing Product Maps for Purchase Decision Making Using AHP

We previously proposed a method for creating product maps with SOM （Self-Organizing Maps） to be used during purchase decision making. In that study, we first established two class boundaries, which divide the area between the minimum and maximum range of an input feature value into three equal parts. Then, we produced self-organizing product maps using classification data inputs. Finally, we applied our method to five product types and confirmed its effectiveness. In this paper, we propose a method for selecting alternatives from a product map, in which we have located a favorite several examples of selecting alternatives and making decisions using cluster, and/or from a favorite component map. We then show the AHP （Analytic Hierarchy Process）.

The Testing Intelligence System Based on Factor Models and Self-Organizing Feature Maps

Presented is a new testing system based on using the factor models and self-organizing feature maps as well as the method of filtering undesirable environment influence. Testing process is described by the factor model with simplex structure, which represents the influences of genetics and environmental factors on the observed parameters - the answers to the questions of the test subjects in one case and for the time, which is spent on responding to each test question to another. The Monte Carlo method is applied to get sufficient samples for training self-organizing feature maps, which are used to estimate model goodness-of-fit measures and, consequently, ability level. A prototype of the system is implemented using the Raven＇s Progressive Matrices （Advanced Progressive Matrices） - an intelligence test of abstract reasoning. Elimination of environment influence results is performed by comparing the observed and predicted answers to the test tasks using the Kalman filter, which is adapted to solve the problem. The testing procedure is optimized by reducing the number of tasks using the distribution of measures to belong to different ability levels after performing each test task provided the required level of conclusion reliability is obtained.

Fault Diagnosis in Chemical Process Based on Self-organizing Map Integrated with Fisher Discriminant Analysis

Fault diagnosis and monitoring are very important for complex chemical process. There are numerous methods that have been studied in this field, in which the effective visualization method is still challenging. In order to get a better visualization effect, a novel fault diagnosis method which combines self-organizing map (SOM) with Fisher discriminant analysis (FDA) is proposed. FDA can reduce the dimension of the data in terms of maximizing the separability of the classes. After feature extraction by FDA, SOM can distinguish the different states on the output map clearly and it can also be employed to monitor abnormal states. Tennessee Eastman (TE) process is employed to illustrate the fault diagnosis and monitoring performance of the proposed method. The result shows that the SOM integrated with FDA method is efficient and capable for real-time monitoring and fault diagnosis in complex chemical process.

Data-Driven Microstructure and Microhardness Design in Additive Manufacturing Using a Self-Organizing Map

To design microstructure and microhardness in the additive manufacturing(AM)of nickel(Ni)-based superalloys,the present work develops a novel data-driven approach that combines physics-based models,experimental measurements,and a data-mining method.The simulation is based on a computational thermal-fluid dynamics(CtFD)model,which can obtain thermal behavior,solidification parameters such as cooling rate,and the dilution of solidified clad.Based on the computed thermal information,dendrite arm spacing and microhardness are estimated using well-tested mechanistic models.Experimental microstructure and microhardness are determined and compared with the simulated values for validation.To visualize process-structure-properties(PSPs)linkages,the simulation and experimental datasets are input to a data-mining model-a self-organizing map(SOM).The design windows of the process parameters under multiple objectives can be obtained from the visualized maps.The proposed approaches can be utilized in AM and other data-intensive processes.Data-driven linkages between process,structure,and properties have the potential to benefit online process monitoring control in order to derive an ideal microstructure and mechanical properties.

Patterns of upper layer circulation variability in the South China Sea from satellite altimetry using the self-organizing map

Patterns of the South China Sea (SCS) circulation variability are extracted from merged satellite altimetry data from October 1992 through August 2004 by using the self-organizing map (SOM). The annual cycle, seasonal and inter-annual variations of the SCS surface circulation are identified through the evolution of the characteristic circulation patterns.The annual cycle of the SCS general circulation patterns is described as a change between two opposite basin-scale SW-NE oriented gyres embedded with eddies: low sea surface height anomaly (SSHA) (cyclonic) in winter and high SSHA (anticyclonic) in summer half year. The transition starts from July—August (January—February) with a high (low) SSHA tongue east of Vietnam around 12°~14° N, which develops into a big anticyclonic (cyclonic) gyre while moving eastward to the deep basin. During the transitions, a dipole structure, cyclonic (anticyclonic) in the north and anticyclonic (cyclonic) in the south, may be formed southeast off Vietnam with a strong zonal jet around 10°~12° N. The seasonal variation is modulated by the interannual variations. Besides the strong 1997/1998 event in response to the peak Pacific El Nio in 1997, the overall SCS sea level is found to have a significant rise during 1999~2001, however, in summer 2004 the overall SCS sea level is lower and the basin-wide anticyclonic gyre becomes weaker than the other years.

Outlier Detection in Near Infra-Red Spectra with Self-Organizing Map

A new method to detect multiple outliers in multivariate data is proposed. It is a combination of minimum subsets, resampling and self-organizing map (SOM) algorithm introduced by Kohonen,which provides a robust way with neural network. In this method, the number and organization of the neurons are selected by the characteristics of the spectra, e.g., the spectra data are often changed linearly with the concentration of the components and are often measured repeatedly, etc. So the spatial distribution of the neurons can be arranged by this characteristic. With this method, all the outliers in the spectra can be detected, which cannot be solved by the traditional method, and the speed of computation is higher than that of the traditional neural network method. The results of the simulation and the experiment show that this method is simple, effective, intuitionistic and all the outliers in the spectra can be detected in a short time. It is useful when associated with the regression model in the near infra-red research.

Fault diagnosis and process monitoring using a statistical pattern framework based on a self-organizing map

A multivariate method for fault diagnosis and process monitoring is proposed. This technique is based on a statistical pattern(SP) framework integrated with a self-organizing map(SOM). An SP-based SOM is used as a classifier to distinguish various states on the output map, which can visually monitor abnormal states. A case study of the Tennessee Eastman(TE) process is presented to demonstrate the fault diagnosis and process monitoring performance of the proposed method. Results show that the SP-based SOM method is a visual tool for real-time monitoring and fault diagnosis that can be used in complex chemical processes.Compared with other SOM-based methods, the proposed method can more efficiently monitor and diagnose faults.

Dimensionality Reduction Using Optimized Self-Organized Map Technique for Hyperspectral Image Classification

The high dimensionalhyperspectral image classification is a challenging task due to the spectral feature vectors.The high correlation between these features and the noises greatly affects the classification performances.To overcome this,dimensionality reduction techniques are widely used.Traditional image processing applications recently propose numerous deep learning models.However,in hyperspectral image classification,the features of deep learning models are less explored.Thus,for efficient hyperspectral image classification,a depth-wise convolutional neural network is presented in this research work.To handle the dimensionality issue in the classification process,an optimized self-organized map model is employed using a water strider optimization algorithm.The network parameters of the self-organized map are optimized by the water strider optimization which reduces the dimensionality issues and enhances the classification performances.Standard datasets such as Indian Pines and the University of Pavia(UP)are considered for experimental analysis.Existing dimensionality reduction methods like Enhanced Hybrid-Graph Discriminant Learning(EHGDL),local geometric structure Fisher analysis(LGSFA),Discriminant Hyper-Laplacian projection(DHLP),Group-based tensor model(GBTM),and Lower rank tensor approximation(LRTA)methods are compared with proposed optimized SOM model.Results confirm the superior performance of the proposed model of 98.22%accuracy for the Indian pines dataset and 98.21%accuracy for the University of Pavia dataset over the existing maximum likelihood classifier,and Support vector machine(SVM).