Two-level k-means clustering algorithm for k-tau relationship establishment and linear-time classification

Partitional clustering algorithms, which partition the dataset into a pre-defined number of clusters, can be broadly classified into two types: algorithms which explicitly take the number of clusters as input and algorithms that take the expected size of a cluster as input. In this paper, we propose a variant of the k-means algorithm and prove that it is more efficient than standard k-means algorithms. An important contribution of this paper is the establishment of a relation between the number of clusters and the size of the clusters in a dataset through the analysis of our algorithm. We also demonstrate that the integration of this algorithm as a pre-processing step in classification algorithms reduces their running-time complexity.

Classification of Indian meteorological stations using cluster and fuzzy cluster analysis, and Kohonen artificial neural networks

The present study deals with the application of cluster analysis, Fuzzy Cluster Analysis (FCA) and Kohonen Artificial Neural Networks (KANN) methods for classification of 159 meteorological stations in India into meteorologically homogeneous groups. Eight parameters, namely latitude, longitude, elevation, average temperature, humidity, wind speed, sunshine hours and solar radiation, are considered as the classification criteria for grouping. The optimal number of groups is determined as 14 based on the Davies-Bouldin index approach. It is observed that the FCA approach performed better than the other two methodologies for the present study.

Kernels for large margin time-series classification

In this paper we propose a novel family of kernels for multivariate time-series classification problems. Each time-series is approximated by a linear combination of piecewise polynomial functions in a Reproducing Kernel Hilbert Space by a novel kernel interpolation technique. Using the associated kernel function a large margin classification formulation is proposed which can discriminate between two classes. The formulation leads to kernels, between two multivariate time-series, which can be efficiently computed. The kernels have been successfully applied to writer independent handwritten character recognition.

Molecular phylogeny and biogeography of langurs and leaf monkeys of South Asia (Primates: Colobinae)

The two recently proposed taxonomies of the langurs and leaf monkeys (Subfamily Colobinae) provide different implications to our understanding of the evolution of Nilgiri and purple-faced langurs. Groves (2001) [Groves, C.P., 2001. Primate Taxonomy. Smithsonian Institute Press, Washington], placed Nilgiri and purple-faced langurs in the genus Trachypithecus, thereby suggesting disjunct distribution of the genus Trachypithecus. [Brandon-Jones, D., Eudey, A.A., Geissmann, T., Groves, C.P., Melnick, D.J., Morales, J.C., Shekelle, M., Stewart, C.-B., 2003. Asian primate classification. Int. J. Primatol. 25, 97–162] placed these langurs in the genus Semnopithecus, which suggests convergence of morphological characters in Nilgiri and purple-faced langurs with Trachypithecus. To test these scenarios, we sequenced and analyzed the mitochondrial cytochrome b gene and two nuclear DNA-encoded genes, lysozyme and protamine P1, from a variety of colobine species. All three markers support the clustering of Nilgiri and purple-faced langurs with Hanuman langur (Semnopithecus), while leaf monkeys of Southeast Asian (Trachypithecus) form a distinct clade. The phylogenetic position of capped and golden leaf monkeys is still unresolved. It is likely that this species group might have evolved due to past hybridization between Semnopithecus and Trachypithecus clades.

Urban satellite image classification using biologically inspired techniques

This paper focuses on optimisation algorithms inspired by swarm intelligence for satellite image classification from high resolution satellite multi- spectral images. Amongst the multiple benefits and uses of remote sensing, one of the most important has been its use in solving the problem of land cover mapping. As the frontiers of space technology advance, the knowledge derived from the satellite data has also grown in sophistication. Image classification forms the core of the solution to the land cover mapping problem. No single classifier can prove to satisfactorily classify all the basic land cover classes of an urban region. In both supervised and unsupervised classification methods, the evolutionary algorithms are not exploited to their full potential. This work tackles the land map covering by Ant Colony Optimisation (ACO) and Particle Swarm Optimisation (PSO) which are arguably the most popular algorithms in this category. We present the results of classification techniques using swarm intelligence for the problem of land cover mapping for an urban region. The high resolution Quick-bird data has been used for the experiments.

A model of social work classification in health care

The introduction of casemix funding for Australian acute health care services has challenged Social Work to demonstrate clear reporting mechanisms, demonstrate effective practice and to justify interventions provided. The term 'casemix' is used to describe the mix and type of patients treated by a hospital or other health care services. There is wide acknowledgement that the procedure-based system of Diagnosis Related Groupings (DRGs) is grounded in a medical/illness perspective and is unsatisfactory in describing and predicting the activity of Social Work and other allied health professions in health care service delivery. The National Allied Health Casemix Committee was established in 1991 as the peak body to represent allied health professions in matters related to casemix classification. This Committee has pioneered a nationally consistent, patient-centred information system for allied health. This paper describes the classification systems and codes developed for Social Work, which includes a minimum data set, a classification hierarchy, the set of activity (input) codes and 'indicator for intervention' codes. The advantages and limitations of the system are also discussed.

Impact of riparian land use on stream insects of Kudremukh National Park, Karnataka state, India

The impact of riparian land use on the stream insect communities was studied at Kudremukh National Park located within Western Ghats, a tropical biodiversity hotspot in India. The diversity and community composition of stream insects varied across streams with different riparian land use types. The rarefied family and generic richness was highest in streams with natural semi evergreen forests as riparian vegetation. However, when the streams had human habitations and areca nut plantations as riparian land use type, the rarefied richness was higher than that of streams with natural evergreen forests and grasslands. The streams with scrub lands and iron ore mining as the riparian land use had the lowest rarefied richness. Within a landscape, the streams with the natural riparian vegetation had similar community composition. However, streams with natural grasslands as the riparian vegetation, had low diversity and the community composition was similar to those of paddy fields. We discuss how stream insect assemblages differ due to varied riparian land use patterns, reflecting fundamental alterations in the functioning of stream ecosystems. This understanding is vital to conserve, manage and restore tropical riverine ecosystems.

Charge density based classification of intermolecular interactions in molecular crystals

Evaluation of intermolecular interactions in terms of both experimental and theoretical charge density analyses has produced a unified picture with which to classify strong and weak hydrogen bonds, along with van der Waals interactions, into three regions.

A Framework for Classification of Prokaryotic Protein Kinases

Background:Overwhelming majority of the Serine/Threonine protein kinases identified by gleaning archaeal and eubacterial genomes could not be classified into any of the well known Hanks and Hunter subfamilies of protein kinases. This is owing to the development of Hanks and Hunter classification scheme based on eukaryotic protein kinases which are highly divergent from their prokaryotic homologues. A large dataset of prokaryotic Serine/Threonine protein kinases recognized from genomes of prokaryotes have been used to develop a classification framework for prokaryotic Ser/Thr protein kinases. Methodology/Principal Findings: We have used traditional sequence alignment and phylogenetic approaches and clustered the prokaryotic kinases which represent 72 subfamilies with at least 4 members in each. Such a clustering enables classification of prokaryotic Ser/Thr kinases and it can be used as a framework to classify newly identified prokaryotic Ser/Thr kinases. After series of searches in a comprehensive sequence database we recognized that 38 subfamilies of prokaryotic protein kinases are associated to a specific taxonomic level. For example 4, 6 and 3 subfamilies have been identified that are currently specific to phylum proteobacteria, cyanobacteria and actinobacteria respectively. Similarly subfamilies which are specific to an order, sub-order, class, family and genus have also been identified. In addition to these, we also identify organism-diverse subfamilies. Members of these clusters are from organisms of different taxonomic levels, such as archaea, bacteria, eukaryotes and viruses.Conclusion/Significance: Interestingly, occurrence of several taxonomic level specific subfamilies of prokaryotic kinases contrasts with classification of eukaryotic protein kinases in which most of the popular subfamilies of eukaryotic protein kinases occur diversely in several eukaryotes. Many prokaryotic Ser/Thr kinases exhibit a wide variety of modular organization which indicates a degree of complexity and protein-protein interactions in the signaling pathways in these microbes.

Energy landscape view of phase transitions and slow dynamics in thermotropic liquid crystals

Thermotropic liquid crystals are known to display rich phase behavior on temperature variation. Although the nematic phase is orientationally ordered but translationally disordered, a smectic phase is characterized by the appearance of a partial translational order in addition to a further increase in orientational order. In an attempt to understand the interplay between orientational and translational order in the mesophases that thermotropic liquid crystals typically exhibit upon cooling from the high-temperature isotropic phase, we investigate the potential energy landscapes of a family of model liquid crystalline systems. The configurations of the system corresponding to the local potential energy minima, known as the inherent structures, are determined from computer simulations across the mesophases. We find that the depth of the potential energy minima explored by the system along an isochor grows through the nematic phase as temperature drops in contrast to its insensitivity to temperature in the isotropic and smectic phases. The onset of the growth of the orientational order in the parent phase is found to induce a translational order, resulting in a smectic-like layer in the underlying inherent structures; the inherent structures, surprisingly, never seem to sustain orientational order alone if the parent nematic phase is sandwiched between the high-temperature isotropic phase and the low-temperature smectic phase. The Arrhenius temperature dependence of the orientational relaxation time breaks down near the isotropic-nematic transition. We find that this breakdown occurs at a temperature below which the system explores increasingly deeper potential energy minima.

Comparison of Multiclass SVM Classification Methods to Use in a Supportive System for Distance Relay Coordination

This paper aims at evaluating the methods of multiclass support vector machines (SVMs) for effective use in distance relay coordination. Also, it describes a strategy of supportive systems to aid the conventional protection philosophy in combating situations where protection systems have maloperated and/or information is missing and provide selective and secure coordinations. SVMs have considerable potential as zone classifiers of distance relay coordination. This typically requires a multiclass SVM classifier to effectively analyze/build the underlying concept between reach of different zones and the apparent impedance trajectory during fault. Several methods have been proposed for multiclass classification where typically several binary SVM classifiers are combined together. Some authors have extended binary SVM classification to one-step single optimization operation considering all classes at once. In this paper, one-step multiclass classification, one-against-all, and one-against-one multiclass methods are compared for their performance with respect to accuracy, number of iterations, number of support vectors, training, and testing time. The performance analysis of these three methods is presented on three data sets belonging to training and testing patterns of three supportive systems for a region and part of a network, which is an equivalent 526-bus system of the practical Indian Western grid.

Classification of magnetic resonance brain images using wavelets as input to support vector machine and neural network

In this paper. we propose a novel method using wavelets as input to neural network self-organizing maps and support vector machine for classification of magnetic resonance (MR) images of the human brain. The proposed method classifies MR brain images as either normal or abnormal. We have tested the proposed approach using a dataset of 52 MR brain images. Good classification percentage of more than 94% was achieved using the neural network self-organizing maps (SOM) and 98% front support vector machine. We observed that the classification rate is high for a Support vector machine classifier compared to self-organizing map-based approach.