Radial basis function classifier construction using particle swarm optimisation aided orthogonal forward regression

We develop a particle swarm optimisation (PSO) aided orthogonal forward regression (OFR) approach for constructing radial basis function (RBF) classifiers with tunable nodes. At each stage of the OFR construction process, the centre vector and diagonal covariance matrix of one RBF node is determined efficiently by minimising the leave-one-out (LOO) misclassification rate (MR) using a PSO algorithm. Compared with the state-of-the-art regularisation assisted orthogonal least square algorithm based on the LOO MR for selecting fixednode RBF classifiers, the proposed PSO aided OFR algorithm for constructing tunable-node RBF classifiers offers significant advantages in terms of better generalisation performance and smaller model size as well as imposes lower computational complexity in classifier construction process. Moreover, the proposed algorithm does not have any hyperparameter that requires costly tuning based on cross validation.

Generalized neurofuzzy network modeling algorithms using Bezier-Bernstein polynomial functions and additive decomposition

This paper introduces a new neurofuzzy model construction algorithm for nonlinear dynamic systems based upon basis functions that are Bezier-Bernstein polynomial functions. This paper is generalized in that it copes with n-dimensional inputs by utilising an additive decomposition construction to overcome the curse of dimensionality associated with high n. This new construction algorithm also introduces univariate Bezier-Bernstein polynomial functions for the completeness of the generalized procedure. Like the B-spline expansion based neurofuzzy systems, Bezier-Bernstein polynomial function based neurofuzzy networks hold desirable properties such as nonnegativity of the basis functions, unity of support, and interpretability of basis function as fuzzy membership functions, moreover with the additional advantages of structural parsimony and Delaunay input space partition, essentially overcoming the curse of dimensionality associated with conventional fuzzy and RBF networks. This new modeling network is based on additive decomposition approach together with two separate basis function formation approaches for both univariate and bivariate Bezier-Bernstein polynomial functions used in model construction. The overall network weights are then learnt using conventional least squares methods. Numerical examples are included to demonstrate the effectiveness of this new data based modeling approach.

Givens rotation based fast backward elimination algorithm for RBF neural network pruning

A fast backward elimination algorithm is introduced based on a QR decomposition and Givens transformations to prune radial-basis-function networks. Nodes are sequentially removed using an increment of error variance criterion. The procedure is terminated by using a prediction risk criterion so as to obtain a model structure with good generalisation properties. The algorithm can be used to postprocess radial basis centres selected using a k-means routine and, in this mode, it provides a hybrid supervised centre selection approach.

The study of chiral adsorption systems using synchrotron- based structural and spectroscopic techniques: stereospecific adsorption of serine on Au-modified chiral Cu{531} surfaces

We apply modern synchrotron-based structural techniques to the study of serine adsorbed on the pure andAumodified intrinsically chiral Cu{531} surface. XPS and NEXAFS data in combination with DFT show that on the pure surface both enantiomers adsorb in l4 geometries (with de-protonated b-OH groups) at low coverage and in l3 geometries at saturation coverage. Significantly larger enantiomeric differences are seen for the l4 geometries, which involve substrate bonds of three side groups of the chiral center, i.e. a three-point interaction. The l3 adsorption geometry, where only the carboxylate and amino groups form substrate bonds, leads to smaller but still significant enantiomeric differences, both in geometry and the decomposition behavior. When Cu{531} is modified by the deposition of 1 and 2ML Au the orientations of serine at saturation coverage are significantly different from those on the clean surface. In all cases, however, a l3 bond coordination is found at saturation involving different numbers of Au atoms, which leads to relatively small enantiomeric differences.

Evaluation of empirical mode decomposition for event-related potential analysis

Current methods for estimating event-related potentials (ERPs) assume stationarity of the signal. Empirical Mode Decomposition (EMD) is a data-driven decomposition technique that does not assume stationarity. We evaluated an EMD-based method for estimating the ERP. On simulated data, EMD substantially reduced background EEG while retaining the ERP. EMD-denoised single trials also estimated shape, amplitude, and latency of the ERP better than raw single trials. On experimental data, EMD-denoised trials revealed event-related differences between two conditions (condition A and B) more effectively than trials lowpass filtered at 40 Hz. EMD also revealed event-related differences on both condition A and condition B that were clearer and of longer duration than those revealed by low-pass filtering at 40 Hz. Thus, EMD-based denoising is a promising data-driven, nonstationary method for estimating ERPs and should be investigated further.

Estimating reliability and resolution of probability forecasts through decomposition of the empirical score

References (20)Cited By (1)Export CitationAboutAbstract Proper scoring rules provide a useful means to evaluate probabilistic forecasts. Independent from scoring rules, it has been argued that reliability and resolution are desirable forecast attributes. The mathematical expectation value of the score allows for a decomposition into reliability and resolution related terms, demonstrating a relationship between scoring rules and reliability/resolution. A similar decomposition holds for the empirical (i.e. sample average) score over an archive of forecast–observation pairs. This empirical decomposition though provides a too optimistic estimate of the potential score (i.e. the optimum score which could be obtained through recalibration), showing that a forecast assessment based solely on the empirical resolution and reliability terms will be misleading. The differences between the theoretical and empirical decomposition are investigated, and specific recommendations are given how to obtain better estimators of reliability and resolution in the case of the Brier and Ignorance scoring rule.

Pollen-based continental climate reconstructions at 6 and 21 ka: a global synthesis

Subfossil pollen and plant macrofossil data derived from 14C-dated sediment profiles can provide quantitative information on glacial and interglacial climates. The data allow climate variables related to growing season warmth, winter cold, and plant-available moisture to be reconstructed. Continental-scale reconstructions have been made for the mid-Holocene (MH, around 6 ka) and Last Glacial Maximum (LGM, around 21 ka), allowing comparison with palaeoclimate simulations currently being carried out as part of the fifth Assessment Report (AR5) of the Intergovernmental Panel on Climate Change. The synthesis of the available MH and LGM climate reconstructions and their uncertainties, obtained using modern analogue, regression and model-inversion techniques, is presented for four temperature variables and two moisture variables. Reconstructions of the same variables based on surface-pollen assemblages are shown to be accurate and unbiased. Reconstructed LGM and MH climate anomaly patterns are coherent, consistent between variables, and robust with respect to the choice of technique. They support a conceptual model of the controls of Late Quaternary climate change whereby the first-order effects of orbital variations and greenhouse forcing on the seasonal cycle of temperature are predictably modified by responses of the atmospheric circulation and surface energy balance.

A neurofuzzy classifier for two class problems

A neurofuzzy classifier identification algorithm is introduced for two class problems. The initial fuzzy base construction is based on fuzzy clustering utilizing a Gaussian mixture model (GMM) and the analysis of covariance (ANOVA) decomposition. The expectation maximization (EM) algorithm is applied to determine the parameters of the fuzzy membership functions. Then neurofuzzy model is identified via the supervised subspace orthogonal least square (OLS) algorithm. Finally a logistic regression model is applied to produce the class probability. The effectiveness of the proposed neurofuzzy classifier has been demonstrated using a real data set.

Elastic-net prefiltering for two-class classification

A two-stage linear-in-the-parameter model construction algorithm is proposed aimed at noisy two-class classification problems. The purpose of the first stage is to produce a prefiltered signal that is used as the desired output for the second stage which constructs a sparse linear-in-the-parameter classifier. The prefiltering stage is a two-level process aimed at maximizing a model's generalization capability, in which a new elastic-net model identification algorithm using singular value decomposition is employed at the lower level, and then, two regularization parameters are optimized using a particle-swarm-optimization algorithm at the upper level by minimizing the leave-one-out (LOO) misclassification rate. It is shown that the LOO misclassification rate based on the resultant prefiltered signal can be analytically computed without splitting the data set, and the associated computational cost is minimal due to orthogonality. The second stage of sparse classifier construction is based on orthogonal forward regression with the D-optimality algorithm. Extensive simulations of this approach for noisy data sets illustrate the competitiveness of this approach to classification of noisy data problems.

Solar ultraviolet-B radiation and vitamin D: a cross-sectional population-based study using data from the 2007 to 2009 Canadian Health Measures Survey

Background: Exposure to solar ultraviolet-B (UV-B) radiation is a major source of vitamin D3. Chemistry climate models project decreases in ground-level solar erythemal UV over the current century. It is unclear what impact this will have on vitamin D status at the population level. The purpose of this study was to measure the association between ground-level solar UV-B and serum concentrations of 25-hydroxyvitamin D (25(OH)D) using a secondary analysis of the 2007 to 2009 Canadian Health Measures Survey (CHMS). Methods: Blood samples collected from individuals aged 12 to 79 years sampled across Canada were analyzed for 25(OH)D (n=4,398). Solar UV-B irradiance was calculated for the 15 CHMS collection sites using the Tropospheric Ultraviolet and Visible Radiation Model. Multivariable linear regression was used to evaluate the association between 25(OH)D and solar UV-B adjusted for other predictors and to explore effect modification. Results: Cumulative solar UV-B irradiance averaged over 91 days (91-day UV-B) prior to blood draw correlated significantly with 25(OH)D. Independent of other predictors, a 1 kJ/m 2 increase in 91-day UV-B was associated with a significant 0.5 nmol/L (95% CI 0.3-0.8) increase in mean 25(OH)D (P =0.0001). The relationship was stronger among younger individuals and those spending more time outdoors. Based on current projections of decreases in ground-level solar UV-B, we predict less than a 1 nmol/L decrease in mean 25(OH)D for the population. Conclusions: In Canada, cumulative exposure to ambient solar UV-B has a small but significant association with 25(OH)D concentrations. Public health messages to improve vitamin D status should target safe sun exposure with sunscreen use, and also enhanced dietary and supplemental intake and maintenance of a healthy body weight.

A unified neurofuzzy model for classification

This work proposes a unified neurofuzzy modelling scheme. To begin with, the initial fuzzy base construction method is based on fuzzy clustering utilising a Gaussian mixture model (GMM) combined with the analysis of covariance (ANOVA) decomposition in order to obtain more compact univariate and bivariate membership functions over the subspaces of the input features. The mean and covariance of the Gaussian membership functions are found by the expectation maximisation (EM) algorithm with the merit of revealing the underlying density distribution of system inputs. The resultant set of membership functions forms the basis of the generalised fuzzy model (GFM) inference engine. The model structure and parameters of this neurofuzzy model are identified via the supervised subspace orthogonal least square (OLS) learning. Finally, instead of providing deterministic class label as model output by convention, a logistic regression model is applied to present the classifier’s output, in which the sigmoid type of logistic transfer function scales the outputs of the neurofuzzy model to the class probability. Experimental validation results are presented to demonstrate the effectiveness of the proposed neurofuzzy modelling scheme.

Particle swarm optimisation assisted classification using elastic net prefiltering

A novel two-stage construction algorithm for linear-in-the-parameters classifier is proposed, aiming at noisy two-class classification problems. The purpose of the first stage is to produce a prefiltered signal that is used as the desired output for the second stage to construct a sparse linear-in-the-parameters classifier. For the first stage learning of generating the prefiltered signal, a two-level algorithm is introduced to maximise the model's generalisation capability, in which an elastic net model identification algorithm using singular value decomposition is employed at the lower level while the two regularisation parameters are selected by maximising the Bayesian evidence using a particle swarm optimization algorithm. Analysis is provided to demonstrate how “Occam's razor” is embodied in this approach. The second stage of sparse classifier construction is based on an orthogonal forward regression with the D-optimality algorithm. Extensive experimental results demonstrate that the proposed approach is effective and yields competitive results for noisy data sets.

Association between serotonin 5-HT-2C receptor gene (HTR2C) polymorphisms and obesity- and mental health-related phenotypes in a large population-based cohort

OBJECTIVE: Studies have shown that common single-nucleotide polymorphisms (SNPs) in the serotonin 5-HT-2C receptor (HTR2C) are associated with antipsychotic agent-induced weight gain and the development of behavioural and psychological symptoms. We aimed to analyse whether variation in the HTR2C is associated with obesity- and mental health-related phenotypes in a large population-based cohort. METHOD: Six tagSNPs, which capture all common genetic variation in the HTR2C gene, were genotyped in 4978 men and women from the European Prospective Investigation into Cancer (EPIC)-Norfolk study, an ongoing prospective population-based cohort study in the United Kingdom. To confirm borderline significant associations, the -759C/T SNP (rs3813929) was genotyped in the remaining 16 003 individuals from the EPIC-Norfolk study. We assessed social and psychological circumstances using the Health and Life Experiences Questionnaire. Genmod models were used to test associations between the SNPs and the outcomes. Logistic regression was performed to test for association of SNPs with obesity- and mental health- related phenotypes. RESULTS: Of the six HTR2C SNPs, only the T allele of the -759C/T SNP showed borderline significant associations with higher body mass index (BMI) (0.23 kg m(-2); (95% confidence interval (CI): 0.01-0.44); P=0.051) and increased risk of lifetime major depressive disorder (MDD) (Odds ratio (OR): 1.13 (95% CI: 1.01-1.22), P=0.02). The associations between the -759C/T and BMI and lifetime MDD were independent. As associations only achieved borderline significance, we aimed to validate our findings on the -759C/T SNP in the full EPIC-Norfolk cohort (n=20 981). Although the association with BMI remained borderline significant (beta=0.20 kg m(-2); 95% CI: 0.04-0.44, P=0.09), that with lifetime MDD (OR: 1.01; 95% CI: 0.94-1.09, P=0.73) was not replicated. CONCLUSIONS: Our findings suggest that common HTR2C gene variants are unlikely to have a major role in obesity- and mental health-related traits in the general population.

The stochastic component in choice and regression to the mean

In this article, we illustrate experimentally an important consequence of the stochastic component in choice behaviour which has not been acknowledged so far. Namely, its potential to produce ‘regression to the mean’ (RTM) effects. We employ a novel approach to individual choice under risk, based on repeated multiple-lottery choices (i.e. choices among many lotteries), to show how the high degree of stochastic variability present in individual decisions can distort crucially certain results through RTM effects. We demonstrate the point in the context of a social comparison experiment.

Measurements of stratospheric ozone at a mid-latitude observing station Valentia, Ireland (51.94° N, 10.25° W), using ground-based and ozonesonde observations from 1994 to 2009

Sixteen years (1994 – 2009) of ozone profiling by ozonesondes at Valentia Meteorological and Geophysical Observatory, Ireland (51.94° N, 10.23° W) along with a co-located MkIV Brewer spectrophotometer for the period 1993–2009 are analyzed. Simple and multiple linear regression methods are used to infer the recent trend, if any, in stratospheric column ozone over the station. The decadal trend from 1994 to 2010 is also calculated from the monthly mean data of Brewer and column ozone data derived from satellite observations. Both of these show a 1.5 % increase per decade during this period with an uncertainty of about ±0.25 %. Monthly mean data for March show a much stronger trend of ~ 4.8 % increase per decade for both ozonesonde and Brewer data. The ozone profile is divided between three vertical slots of 0–15 km, 15–26 km, and 26 km to the top of the atmosphere and a 11-year running average is calculated. Ozone values for the month of March only are observed to increase at each level with a maximum change of +9.2 ± 3.2 % per decade (between years 1994 and 2009) being observed in the vertical region from 15 to 26 km. In the tropospheric region from 0 to 15 km, the trend is positive but with a poor statistical significance. However, for the top level of above 26 km the trend is significantly positive at about 4 % per decade. The March integrated ozonesonde column ozone during this period is found to increase at a rate of ~6.6 % per decade compared with the Brewer and satellite positive trends of ~5 % per decade.