Exploring ant-based algorithms for gene expression data analysis

Objective: Recently, much research has been proposed using nature inspired algorithms to perform complex machine learning tasks. Ant colony optimization (ACO) is one such algorithm based on swarm intelligence and is derived from a model inspired by the collective foraging behavior of ants. Taking advantage of the ACO in traits such as self-organization and robustness, this paper investigates ant-based algorithms for gene expression data clustering and associative classification. Methods and material: An ant-based clustering (Ant-C) and an ant-based association rule mining (Ant-ARM) algorithms are proposed for gene expression data analysis. The proposed algorithms make use of the natural behavior of ants such as cooperation and adaptation to allow for a flexible robust search for a good candidate solution. Results: Ant-C has been tested on the three datasets selected from the Stanford Genomic Resource Database and achieved relatively high accuracy compared to other classical clustering methods. Ant-ARM has been tested on the acute lymphoblastic leukemia (ALL)/acute myeloid leukemia (AML) dataset and generated about 30 classification rules with high accuracy. Conclusions: Ant-C can generate optimal number of clusters without incorporating any other algorithms such as K-means or agglomerative hierarchical clustering. For associative classification, while a few of the well-known algorithms such as Apriori, FP-growth and Magnum Opus are unable to mine any association rules from the ALL/AML dataset within a reasonable period of time, Ant-ARM is able to extract associative classification rules.

Practical methods of tracking of nonstationary time series applied to real-world data

In this paper, we discuss some practical implications for implementing adaptable network algorithms applied to non-stationary time series problems. Two real world data sets, containing electricity load demands and foreign exchange market prices, are used to test several different methods, ranging from linear models with fixed parameters, to non-linear models which adapt both parameters and model order on-line. Training with the extended Kalman filter, we demonstrate that the dynamic model-order increment procedure of the resource allocating RBF network (RAN) is highly sensitive to the parameters of the novelty criterion. We investigate the use of system noise for increasing the plasticity of the Kalman filter training algorithm, and discuss the consequences for on-line model order selection. The results of our experiments show that there are advantages to be gained in tracking real world non-stationary data through the use of more complex adaptive models.

The deoxidation of steel with simple and complex deoxidisers

The deoxidation of steel with complex deoxidisers was studied at 1550°C and compared with silicon, aluminium and silicon/aluminium alloys as standards. The deoxidation alloy systems, Ca/Si/Al, Mg/Si/Al and Mn/Si/Al, were chosen for the low liquidus temperatures of many of their oxide mixtures and the potential deoxidising power of their constituent elements. Product separation rates and compositional relationships following deoxidation were examined. Silicon/aluminium alloy deoxidation resulted in the product compositions and residual oxygen contents expected from equilibrium and stoichiometric considerations, but with the Ca/Si/Al and Mg/Si/Al alloys the volatility of calcium and magnesium prevented them participating in the final solute equilibrium, despite their reported solubility in liquid iron. Electron-probe microanalysis of the products showed various concentrations of lime and magnesia, possibly resulting from reaction between the metal vapours and dissolved oxygen.The consequent reduction of silica activity in the products due to the presence of CaO and hgO produced an indirect effect of calcium and magnesium on the residual oxygen content. Product separation rates, indicated by vacuum fusion analyses, were not significantly influenced by calcium and magnesium but the rapid separation of products having a high Al2O3Si02 ratio was confirmed. Manganese participated in deoxidation, when present either as an alloying element in the steel or as a deoxidation alloy constituent. The compositions of initial oxide products were related to deoxidation alloy compositions. Separated products which were not alumina saturated, dissolved crucible material to achieve saturation. The melt equilibrated with this slag and crucible by diffusion to determine the residual oxygen content. MnO and SiO2 activities were calculated, and the approximate values of MnO deduced for the compositions obtained. Separation rates were greater for products of high interfacial tension. The rates calculated from a model based on Stoke's Law, showed qualitative agreement with experimental data when corrected for coalescence effects.

A tale of two analyses: the use of archived qualitative data

This article provides a unique contribution to the debates about archived qualitative data by drawing on two uses of the same data - British Migrants in Spain: the Extent and Nature of Social Integration, 2003-2005 - by Jones (2009) and Oliver and O'Reilly (2010), both of which utilise Bourdieu's concepts analytically and produce broadly similar findings. We argue that whilst the insights and experiences of those researchers directly involved in data collection are important resources for developing contextual knowledge used in data analysis, other kinds of critical distance can also facilitate credible data use. We therefore challenge the assumption that the idiosyncratic relationship between context, reflexivity and interpretation limits the future use of data. Moreover, regardless of the complex genealogy of the data itself, given the number of contingencies shaping the qualitative research process and thus the potential for partial or inaccurate interpretation, contextual familiarity need not be privileged over other aspects of qualitative praxis such as sustained theoretical insight, sociological imagination and methodological rigour. © Sociological Research Online, 1996-2012.

A data structure for improved gp analysis via efficient computation and visualisation of population measures

Population measures for genetic programs are defined and analysed in an attempt to better understand the behaviour of genetic programming. Some measures are simple, but do not provide sufficient insight. The more meaningful ones are complex and take extra computation time. Here we present a unified view on the computation of population measures through an information hypertree (iTree). The iTree allows for a unified and efficient calculation of population measures via a basic tree traversal. © Springer-Verlag 2004.

Crystal structure and magnetic behaviour of a five-coordinate iron(III) complex of pyridoxal-4-methylthiosemicarbazone

The crystal structure and magnetic properties of a penta-coordinate iron(III) complex of pyridoxal-4-methylthiosemicarbazone, [Fe(Hmthpy)Cl](CHCHSO), are reported. The synthesised ligand and the metal complex were characterised by spectroscopic methods (H NMR, IR, and mass spectroscopy), elemental analysis, and single crystal X-ray diffraction. The complex crystallises as dark brown microcrystals. The crystal data determined at 100(1) K revealed a triclinic system, space group P over(1, ¯) (Z = 2). The ONSCl geometry around the iron(III) atom is intermediate between trigonal bipyramidal and square pyramidal (t = 0.40). The temperature dependence of the magnetic susceptibility (5-300 K) is consistent with a high spin Fe(III) ion (S = 5/2) exhibiting zero-field splitting. Interpretation of these data yielded: D = 0.34(1) cm and g = 2.078(3). © 2007 Elsevier B.V. All rights reserved.

Semisupervised learning of hierarchical latent trait models for data visualization

Recently, we have developed the hierarchical Generative Topographic Mapping (HGTM), an interactive method for visualization of large high-dimensional real-valued data sets. In this paper, we propose a more general visualization system by extending HGTM in three ways, which allows the user to visualize a wider range of data sets and better support the model development process. 1) We integrate HGTM with noise models from the exponential family of distributions. The basic building block is the Latent Trait Model (LTM). This enables us to visualize data of inherently discrete nature, e.g., collections of documents, in a hierarchical manner. 2) We give the user a choice of initializing the child plots of the current plot in either interactive, or automatic mode. In the interactive mode, the user selects "regions of interest," whereas in the automatic mode, an unsupervised minimum message length (MML)-inspired construction of a mixture of LTMs is employed. The unsupervised construction is particularly useful when high-level plots are covered with dense clusters of highly overlapping data projections, making it difficult to use the interactive mode. Such a situation often arises when visualizing large data sets. 3) We derive general formulas for magnification factors in latent trait models. Magnification factors are a useful tool to improve our understanding of the visualization plots, since they can highlight the boundaries between data clusters. We illustrate our approach on a toy example and evaluate it on three more complex real data sets. © 2005 IEEE.

MEG and complex systems

MEG beamformer algorithms work by making the assumption that correlated and spatially distinct local field potentials do not develop in the human brain. Despite this assumption, images produced by such algorithms concur with those from other non-invasive and invasive estimates of brain function. In this paper we set out to develop a method that could be applied to raw MEG data to explicitly test his assumption. We show that a promax rotation of MEG channel data can be used as an approximate estimator of the number of spatially distinct correlated sources in any frequency band.

Evaluating the evaluator:modelling systematic data analysis strategies for software selection

Developers of interactive software are confronted by an increasing variety of software tools to help engineer the interactive aspects of software applications. Not only do these tools fall into different categories in terms of functionality, but within each category there is a growing number of competing tools with similar, although not identical, features. Choice of user interface development tool (UIDT) is therefore becoming increasingly complex.

IL-8 released constitutively by primary bronchial epithelial cells in culture forms an inactive complex with secretory component

The bronchial epithelium is a source of both α and β chemokines and, uniquely, of secretory component (SC), the extracellular ligand-binding domain of the polymeric IgA receptor. Ig superfamily relatives of SC, such as IgG and α2-macroglobulin, bind IL-8. Therefore, we tested the hypothesis that SC binds IL-8, modifying its activity as a neutrophil chemoattractant. Primary bronchial epithelial cells were cultured under conditions to optimize SC synthesis. The chemokines IL-8, epithelial neutrophil-activating peptide-78, growth-related oncogene α, and RANTES were released constitutively by epithelial cells from both normal and asthmatic donors and detected in high m.w. complexes with SC. There were no qualitative differences in the production of SC-chemokine complexes by epithelial cells from normal or asthmatic donors, and in all cases this was the only form of chemokine detected. SC contains 15% N-linked carbohydrate, and complete deglycosylation with peptide N-glycosidase F abolished IL-8 binding. In micro-Boyden chamber assays, no IL-8-dependent neutrophil chemotactic responses to epithelial culture supernatants could be demonstrated. SC dose-dependently (IC50 ∼0.3 nM) inhibited the neutrophil chemotactic response to rIL-8 (10 nM) in micro-Boyden chamber assays and also inhibited IL-8-mediated neutrophil transendothelial migration. SC inhibited the binding of IL-8 to nonspecific binding sites on polycarbonate filters and endothelial cell monolayers, and therefore the formation of haptotactic gradients, without effects on IL-8 binding to specific receptors on neutrophils. The data indicate that in the airways IL-8 may be solubilized and inactivated by binding to SC

Development of Procedures of Recognition of Objects with Usage Multisensor Ontology Controlled Instrumental Complex

The ontological approach to structuring knowledge and the description of data domain of knowledge is considered. It is described tool ontology-controlled complex for research and developments of sensor systems. Some approaches to solution most frequently meeting tasks are considered for creation of the recognition procedures.

CEPBen:a benchmark for complex event processing systems

Complex Event processing (CEP) has emerged over the last ten years. CEP systems are outstanding in processing large amount of data and responding in a timely fashion. While CEP applications are fast growing, performance management in this area has not gain much attention. It is critical to meet the promised level of service for both system designers and users. In this paper, we present a benchmark for complex event processing systems: CEPBen. The CEPBen benchmark is designed to evaluate CEP functional behaviours, i.e., filtering, transformation and event pattern detection and provides a novel methodology of evaluating the performance of CEP systems. A performance study by running the CEPBen on Esper CEP engine is described and discussed. The results obtained from performance tests demonstrate the influences of CEP functional behaviours on the system performance. © 2014 Springer International Publishing Switzerland.

Intellect Sensing of Neural Network that Trained to Classify Complex Signals

An experimental comparison of information features used by neural network is performed. The sensing method was used. Suboptimal classifier agreeable to the gaussian model of the training data was used as a probe. Neural nets with architectures of perceptron and feedforward net with one hidden layer were used. The experiments were carried out with spatial ultrasonic data, which are used for car’s passenger safety system neural controller learning. In this paper we show that a neural network doesn’t fully make use of gaussian components, which are first two moment coefficients of probability distribution. On the contrary, the network can find more complicated regularities inside data vectors and thus shows better results than suboptimal classifier. The parallel connection of suboptimal classifier improves work of modular neural network whereas its connection to the network input improves the specialization effect during training.

The Complex Unified Evolutionary Approach to the Creation of the Multilevel Distributed Control System of a Gas-transport Company

The objects of a large-scale gas-transport company (GTC) suggest a complex unified evolutionary approach, which covers basic building concepts, up-to-date technologies, models, methods and means that are used in the phases of design, adoption, maintenance and development of the multilevel automated distributed control systems (ADCS).. As a single methodological basis of the suggested approach three basic Concepts, which contain the basic methodological principles and conceptual provisions on the creation of distributed control systems, were worked out: systems of the lower level (ACS of the technological processes based on up-to-date SCADA), of the middle level (ACS of the operative-dispatch production control based on MES-systems) and of the high level (business process control on the basis of complex automated systems ERP).

Modeling of Effective Process of Network Maintaining Based on Statistical Data

This paper is dedicated to modelling of network maintaining based on live example – maintaining ATM banking network, where any problems are mean money loss. A full analysis is made in order to estimate valuable and not-valuable parameters based on complex analysis of available data. Correlation analysis helps to estimate provided data and to produce a complex solution of increasing network maintaining effectiveness.