Traditional Cluster Analysis Research Articles

Insights into the implementation of synoptic weather-type classification using self-organizing maps: an Australian case study

The two-fold utility (data projection and cluster analysis) of a two-phase batch self-organizing map (SOM) procedure (CP2) has been previously explored using the NCEP/NCAR geopotential height data for east Australia. That study focused on examining the performance of CP2 in comparison with a traditional cluster analysis procedure, CP1, for the purpose of synoptic typing. The present paper provides additional documentation on the implementation of CP2 for the same region, with broader considerations on the effect of SOM map size, seasonality, data standardization and the choice of neighbourhood functions. A total of 215 SOMs (classifications) were trained through CP2 with various data processing and parameter settings. The examination of these SOMs shows that the two-fold utility of CP2 leads to supplementary visualization of the dominant synoptic patterns over the study region. For SOMs of the same map size (i.e. number of synoptic types), cluster analysis via CP2 provides data groupings with relatively high accuracy and large separation but reduced level of pattern self-organization, while data projection via CP2 tends to create data groupings with a high level of pattern self-organization but reduced accuracy and separation. The choice of map size affects the accuracy, separation and self-organization of data groupings. As a compromise, a map size of 10–20 for cluster analysis and 20–30 for data projection is recommended for the study region. To account for the seasonality and latitudinal heterogeneity in the activity of synoptic systems, a relatively larger SOM size is needed to capture typical synoptic features prevailing in different seasons. Data standardization helps to provide a relatively balanced representation between larger-scale synoptic systems (e.g. anticyclones) and smaller-scale synoptic features (e.g. thermal lows), and also improves the level of pattern self-organization on the SOM across seasons. The additional documentation in this paper encourages a wider application of CP2 in environmental research.

A Clustering Algorithm Based on Improved Particle Swarm Optimization

K-means algorithm is a traditional cluster analysis method, has the characteristics of simple ideas and algorithms, and thus become one of the commonly used methods of cluster analysis. However, the K-means algorithm classification results are too dependent on the choice of the initial cluster centers for some initial value, the algorithm may converge in general suboptimal solutions. Analysis of the K-means algorithm and particle swarm optimization based on a clustering algorithm based on improved particle swarm algorithm. The algorithm local search ability of the K-means algorithm and the global search ability of particle swarm optimization, local search ability to improve the K-means algorithm to accelerate the convergence speed effectively prevent the occurrence of the phenomenon of precocious puberty. The experiments show that the clustering algorithm has better convergence effect.

A MULTI STAGE HEURISTIC FOR MANUFACTURING CELL FORMATION

Cellular manufacturing (CM) is a modern manufacturi ng concept based on the Group technology (GT) philo sophy. CM enables a manufacturing unit to enhance its productivity subs tantially through improved utilization of resources . In Cellular manufacturing system design process the task of solving cell form ation problem (machine ‐component grouping) is the central issue. Cell formation problem is considered as complex one due to its combinatorial nature. Research interest is g rowing towards development of novel solution methodologies for the purpose of cell formation. Even though numerous ap proaches have been implemented by researchers for the task of machine- component grouping, the effectiveness of multi stag e solution procedures based on the integration of traditional and non-tra ditional optimization techniques needs to be invest igated adequately. This paper demonstrates the development and implementati on of a multi stage heuristic based on the integrat ion of traditional cluster analysis technique and non- traditional simulated a nnealing technique for standard cell formation. The proposed heuristic is tested using a number of binary and non-binary data sets and computational results are presented. Resul ts indicate that proposed heuristic is capable of generating promising qualit y solutions when compared with a few alternate cell forming approaches.

An Earthquake Clustering Method Based on Soft Distance Calculations

The traditional cluster analysis method based on the true distance is not conducive to the accurate calculation of earthquake different fault rupture propagation and healing rate. This paper proposed and gave a new clustering method based on soft distance calculations. The clustering process based on soft distance calculations, the calculation method for soft distances and the specific clustering algorithm based on soft distances are given. For the real sample points of strong earthquake as a data source, we use this clustering method and other traditional clustering methods to cluster and analysis the data source, and analysis results have showed that the clustering method obtained the same cluster center with the earth stress field evolution, so this method has objective truth. The cluster analysis method for the earthquake fault zones in the accurate calculation of the next strong earthquake provides a good basis for the calculations.

Clustering Algorithm Based on Improved Particle Swarm Algorithm

K-means algorithm is a traditional cluster analysis method, has the characteristics of simple ideas and algorithms, and thus become one of the commonly used methods of cluster analysis. However, the K-means algorithm classification results are too dependent on the choice of the initial cluster centers for some initial value, the algorithm may converge in general suboptimal solutions. Analysis of the K-means algorithm and particle swarm optimization based on a clustering algorithm based on improved particle swarm algorithm. The algorithm local search ability of the K-means algorithm and the global search ability of particle swarm optimization, local search ability to improve the K-means algorithm to accelerate the convergence speed effectively prevent the occurrence of the phenomenon of precocious puberty. The experiments show that the clustering algorithm has better convergence effect.

A Strategic Group Analysis of Financial Holding Industry in Taiwan - Structural Relationship and Dynamic Development

This study aims to conduct analysis of the dynamic development of the strategic groups of fourteen Financial Holding Companies (FHCs) in Taiwan in terms of diversification, entry barriers and asset turnover by means of the MDPREF Model of MDS during the 2008-2010 period. The findings show that these FHCs are divided into Leader Group, Follower Group and Laggard Group. Time-axis vector was further applied to the longitudinal analysis of FHCs to have a clear understanding of their annual strategic choices. The study differs from the traditional cluster analysis that cannot show the results in two-dimensional graphics.

Analyzing phonetic variation in the traditional English dialects: Simultaneously clustering dialects and phonetic features

This study explores the linguistic application of bipartite spectral graph partitioning, a graph-theoretic technique that simultaneously identifies clusters of similar localities as well as clusters of features characteristic of those localities. We compare the results using this approach with previously published results on the same dataset using cluster and principal component analysis (Shackleton 2007). Although the results of the spectral partitioning method and Shackleton's approach overlap to a broad extent, the analyses offer complementary insights into the data. The traditional cluster analysis detects some clusters that are not identified by the spectral partitioning analysis, whereas the reverse also occurs. Similarly, the principal component analysis and the spectral partitioning analysis detect many overlapping but also some different linguistic variants. The main benefit of the bipartite spectral graph partitioning method over the alternative approaches remains its ability to simultaneously identify sensible geographical clusters of localities with their corresponding linguistic features.

Spatial Cluster Analysis by the Bin-Packing Problem and DNA Computing Technique

Spatial cluster analysis is an important data mining task. Typical techniques include CLARANS, density- and gravity-based clustering, and other algorithms based on traditional von Neumann's computing architecture. The purpose of this paper is to propose a technique for spatial cluster analysis based on sticker systems of DNA computing. We will adopt the Bin-Packing Problem idea and then design algorithms of sticker programming. The proposed technique has a better time complexity. In the case when only the intracluster dissimilarity is taken into account, this time complexity is polynomial in the amount of data points, which reduces the NP-completeness nature of spatial cluster analysis. The new technique provides an alternative method for traditional cluster analysis.

Traditional Chinese Medicine Clustering Analysis Based on Near-Infrared Spectroscopy

The Chinese materia medica is classified into several categories according to the major treatment function by experience in the traditional Chinese Medicine. This classification should have material basis. As the medicines have so many ingredients, it is hard to only use the chemical methods and biochemistry methods to analyze the Chinese materia medica. Near infrared spectroscopy is a nondestructive testing method and it can provide structural information of organic molecules which are the main effective components of the Chinese materia medica. We firstly get the NIR spectrum of some Chinese medicine, and then use LSA technology to process the data and cluster them. The result is quite coincident with the Chinese traditional medicine classification.

Computational gene network analysis reveals TNF-induced angiogenesis

BackgroundTNF (Tumor Necrosis Factor-α) induces HUVEC (Human Umbilical Vein Endothelial Cells) to proliferate and form new blood vessels. This TNF-induced angiogenesis plays a key role in cancer and rheumatic disease. However, the molecular system that underlies TNF-induced angiogenesis is largely unknown.MethodsWe analyzed the gene expression changes stimulated by TNF in HUVEC over a time course using microarrays to reveal the molecular system underlying TNF-induced angiogenesis. Traditional k-means clustering analysis was performed to identify informative temporal gene expression patterns buried in the time course data. Functional enrichment analysis using DAVID was then performed for each cluster. The genes that belonged to informative clusters were then used as the input for gene network analysis using a Bayesian network and nonparametric regression method. Based on this TNF-induced gene network, we searched for sub-networks related to angiogenesis by integrating existing biological knowledge.Resultsk-means clustering of the TNF stimulated time course microarray gene expression data, followed by functional enrichment analysis identified three biologically informative clusters related to apoptosis, cellular proliferation and angiogenesis. These three clusters included 648 genes in total, which were used to estimate dynamic Bayesian networks. Based on the estimated TNF-induced gene networks, we hypothesized that a sub-network including IL6 and IL8 inhibits apoptosis and promotes TNF-induced angiogenesis. More particularly, IL6 promotes TNF-induced angiogenesis by inducing NF-κB and IL8, which are strong cell growth factors.ConclusionsComputational gene network analysis revealed a novel molecular system that may play an important role in the TNF-induced angiogenesis seen in cancer and rheumatic disease. This analysis suggests that Bayesian network analysis linked to functional annotation may be a powerful tool to provide insight into disease.

Comparison of genetic diversity structure analyses of SSR molecular marker data within apple (Malus×domestica) genetic resources

The aim of this study was to compare traditional hierarchical clustering techniques and principal coordinate analysis (PCoA) with the model-based Bayesian cluster analyses in relation to subpopulation differentiation based on breeding history and geographical origin of apple (Malus×domestica Borkh.) cultivars and landraces. We presented the use of a set of 10 microsatellite (SSR) loci for genetic diversity structure analyses of 273 apple accessions from national genetic resources. These SSR loci yielded a total of 113 polymorphic SSR alleles, with 5-18 alleles per locus. SSR molecular data were successfully used in binary and allelic input format for all genetic diversity analyses, but allelic molecular data did not reveal reliable results with the NTSYS-pc and BAPS softwares. A traditional cluster analysis still provided an easy and effective way for determining genetic diversity structure in the apple germplasm collection. A model-based Bayesian analysis also provided the clustering results in accordance to traditional cluster analysis, but the analyses were distorted by the presence of a dominant group of apple genetic resources owing to the narrow origin of the apple genome. PCoA confirmed that there were no noticeable differences in genetic diversity structure of apple genetic resources during the breeding history. The results of our analyses are useful in the context of enhancing apple collection management, sampling of core collections, and improving breeding processes.

Research on Clustering Analysis of Big Data

Data quantity of Big Data was too big to be processed with traditional clustering analysis technologies. Time consuming was long, problem of computability existed with traditional technologies. Having analyzed on k-means clustering algorithm, a new algorithm was proposed. Parallelizing part of k-means was found. The algorithm was improved with the method of redesigning flow with MapReduce framework. Problems mentioned above were solved. Experiments show that new algorithm is feasible and effective.

Quantitative log interpretation and uncertainty propagation of petrophysical properties and facies classification from rock-physics modeling and formation evaluation analysis

Formation evaluation analysis, rock-physics models, and log-facies classification are powerful tools to link the physical properties measured at wells with petrophysical, elastic, and seismic properties. However, this link can be affected by several sources of uncertainty. We proposed a complete statistical workflow for obtaining petrophysical properties at the well location and the corresponding log-facies classification. This methodology is based on traditional formation evaluation models and cluster analysis techniques, but it introduces a full Monte Carlo approach to account for uncertainty evaluation. The workflow includes rock-physics models in log-facies classification to preserve the link between petrophysical properties, elastic properties, and facies. The use of rock-physics model predictions guarantees obtaining a consistent set of well-log data that can be used both to calibrate the usual physical models used in seismic reservoir characterization and to condition reservoir models. The final output is the set of petrophysical curves with the associated uncertainty, the profile of the facies probabilities, and the entropy, or degree of confusion, related to the most probable facies profile. The full statistical approach allows us to propagate the uncertainty from data measured at the well location to the estimated petrophysical curves and facies profiles. We applied the proposed methodology to two different well-log studies to determine its applicability, the advantages of the new integrated approach, and the value of uncertainty analysis.

Relaciones de gobernanza e innovación en la cadena de valor: nuevos paradigmas de competividad

As a result of the development of emerging countries such as China and Brazil, the traditional competitive paradigm based on the cost benefits and efficiencies related to quality, are not enough nowadays. These paradigms related the classical competitive with the degree of technological innovation and the use of cluster resources by its members. However, the combination of adequate knowledge and relationship marketing management provides a review of current models of competition that goes beyond the traditional cluster analysis. The aim of this study is to analyze the governance structure of the territorial value chain in the Spanish ceramic tile cluster, through the understanding of past and current roles of the various sectors involved in the creation of the value system. Through the case study approach and quantitative methodology, the study explores the paradigm shift where the traditional actors in the value chain are losing their control over their contribution to the value creation system, while new players with a more stable and promising role are emerging.

Comparing dissimilarity frequency distributions and dendrograms of cluster analysis in case of plant variety description matrices

Plant variety descriptions are based on standardized data of morphological characteristics. Intervarietal similarity was calculated by pairwise comparison. Random numbers of 1–9 range were used in the study in order to model clear similarity relations. Similarity groups were defined shorter intervals of random numbers. Dissimilarity values were used for constructing dendrograms and dissimilarity frequency distribution (DFD) curves. DFD curve patterns and dendrograms obtained in cluster analysis were compared on eight model matrices. Dendrograms clearly separated pre-defined similarity groups but created clusters from uncorrelated random number matrices as well. DFD patterns proved to be suitable to detect similarity. The number and shape of peaks and their x axis position provided information on data set variability and similarity relations. DFD patterns clearly indicated the merging process of increased overlap of similarity groups. The new method can complement traditional cluster analysis in evaluation similarity.

Spatial Cluster Analysis by the Adleman‐Lipton DNA Computing Model and Flexible Grids

Spatial cluster analysis is an important data‐mining task. Typical techniques include CLARANS, density‐ and gravity‐based clustering, and other algorithms based on traditional von Neumann’s computing architecture. The purpose of this paper is to propose a technique for spatial cluster analysis based on DNA computing and a grid technique. We will adopt the Adleman‐Lipton model and then design a flexible grid algorithm. Examples are given to show the effect of the algorithm. The new clustering technique provides an alternative for traditional cluster analysis.

Typology of Fashion Product Consumers: Application of Mixture-model Segmentation Analysis

Proper consumer segmentation is receiving more attention from industry professionals as markets become more diverse and consumer-centered. Researchers have recognized the limitations of the traditional cluster analysis technique and this research study analyzes market segmentation using Mixture-model or latent-class segmentation. This study used a questionnaire to determine the characteristics of clothing shoppers using a new technique that proved its superiority over traditional techniques. Questions included items measuring fashion shopping behavior, store choice criteria, apparel consumption styles, price perception by product type, and demographic characteristics. Data were collected from 1074 males and females in their 20s and 30s through an online survey. SPSS 16.0 and Latent GOLD 4.0 were used to analyze the data. The ideal typology of clothing shoppers using the Mixture-model were: `brand loyalty orientated group`, `group of conservative late 30s`, `group of pleasure-emotion early 20s`, `value oriented consumer product with high-income group`, `group of eco/symbol oriented consumer`, and `group of utility/goal oriented male consumer`. This study showed differences in fashion product purchasing behavior by conducting market segmentation for clothing shoppers using the Mixture-model.

The Research on Modular Designing Method of Marine Diesel Engine

Modular design has gradually become an important method to modern design, and the modular method is the key. This paper combines extension theory with clustering analysis methods through establishing matter elements model of marine diesel engine, which is the extension of the traditional extenics and clustering analysis methods, and analysis and evaluate the results. It is the foundation of the modern design of the marine diesel engine.

Farm Landowners' Objectives in Finland: Two Approaches for Owner Classifications

Two landowner classification methods based on owner objectives have been used to describe the heterogeneous group of farmland owners. As accurate information on landowner preferences is essential in policy planning and evaluation of the effects of various policy instruments, there is a need to develop feasible methods for classifying landowners. In this study we applied objective-based classification to Finnish farmland owners. The two classification methods compared in terms of their criterion validity were traditional cluster analysis and latent class analysis. Comparison of the convergent, concurrent, discriminant, and predictive components of criterion validity revealed that latent class analysis, particularly if sociodemographic covariates were included, was a slightly better classification method. Classification of farmland owners according to their ownership objectives was shown to be a relevant predictor of landowner behavior, and thus to provide valuable information for agricultural policymakers.

Determination of genetic structure of germplasm collections: are traditional hierarchical clustering methods appropriate for molecular marker data?

Despite the availability of newer approaches, traditional hierarchical clustering remains very popular in genetic diversity studies in plants. However, little is known about its suitability for molecular marker data. We studied the performance of traditional hierarchical clustering techniques using real and simulated molecular marker data. Our study also compared the performance of traditional hierarchical clustering with model-based clustering (STRUCTURE). We showed that the cophenetic correlation coefficient is directly related to subgroup differentiation and can thus be used as an indicator of the presence of genetically distinct subgroups in germplasm collections. Whereas UPGMA performed well in preserving distances between accessions, Ward excelled in recovering groups. Our results also showed a close similarity between clusters obtained by Ward and by STRUCTURE. Traditional cluster analysis can provide an easy and effective way of determining structure in germplasm collections using molecular marker data, and, the output can be used for sampling core collections or for association studies.Electronic supplementary materialThe online version of this article (doi:10.1007/s00122-011-1576-x) contains supplementary material, which is available to authorized users.