On improving the generalization of SVM classifier

The generalization performance of the SVM classifier depends mainly on the VC dimension and the dimensionality of the data. By reducing the VC dimension of the SVM classifier, its generalization performance is expected to increase. In the present paper, we argue that the VC dimension of SVM classifier can be reduced by applying bootstrapping and dimensionality reduction techniques. Experimental results showed that bootstrapping the original data and bootstrapping the projected (dimensionally reduced) data improved the performance of the SVM classifier.

A game theoretic approach for feature clustering and its application to feature selection

In this paper, we develop a game theoretic approach for clustering features in a learning problem. Feature clustering can serve as an important preprocessing step in many problems such as feature selection, dimensionality reduction, etc. In this approach, we view features as rational players of a coalitional game where they form coalitions (or clusters) among themselves in order to maximize their individual payoffs. We show how Nash Stable Partition (NSP), a well known concept in the coalitional game theory, provides a natural way of clustering features. Through this approach, one can obtain some desirable properties of the clusters by choosing appropriate payoff functions. For a small number of features, the NSP based clustering can be found by solving an integer linear program (ILP). However, for large number of features, the ILP based approach does not scale well and hence we propose a hierarchical approach. Interestingly, a key result that we prove on the equivalence between a k-size NSP of a coalitional game and minimum k-cut of an appropriately constructed graph comes in handy for large scale problems. In this paper, we use feature selection problem (in a classification setting) as a running example to illustrate our approach. We conduct experiments to illustrate the efficacy of our approach.

Crop stage classification of hyperspectral data using unsupervised techniques

The presence of a large number of spectral bands in the hyperspectral images increases the capability to distinguish between various physical structures. However, they suffer from the high dimensionality of the data. Hence, the processing of hyperspectral images is applied in two stages: dimensionality reduction and unsupervised classification techniques. The high dimensionality of the data has been reduced with the help of Principal Component Analysis (PCA). The selected dimensions are classified using Niche Hierarchical Artificial Immune System (NHAIS). The NHAIS combines the splitting method to search for the optimal cluster centers using niching procedure and the merging method is used to group the data points based on majority voting. Results are presented for two hyperspectral images namely EO-1 Hyperion image and Indian pines image. A performance comparison of this proposed hierarchical clustering algorithm with the earlier three unsupervised algorithms is presented. From the results obtained, we deduce that the NHAIS is efficient.

Least squares QR-based decomposition provides an efficient way of computing optimal regularization parameter in photoacoustic tomography

A computationally efficient approach that computes the optimal regularization parameter for the Tikhonov-minimization scheme is developed for photoacoustic imaging. This approach is based on the least squares-QR decomposition which is a well-known dimensionality reduction technique for a large system of equations. It is shown that the proposed framework is effective in terms of quantitative and qualitative reconstructions of initial pressure distribution enabled via finding an optimal regularization parameter. The computational efficiency and performance of the proposed method are shown using a test case of numerical blood vessel phantom, where the initial pressure is exactly known for quantitative comparison. (C) 2013 Society of Photo-Optical Instrumentation Engineers (SPIE)

A NEAR OPTIMAL PROJECTION FOR SPARSE REPRESENTATION BASED CLASSIFICATION

Sparse representation based classification (SRC) is one of the most successful methods that has been developed in recent times for face recognition. Optimal projection for Sparse representation based classification (OPSRC)1] provides a dimensionality reduction map that is supposed to give optimum performance for SRC framework. However, the computational complexity involved in this method is too high. Here, we propose a new projection technique using the data scatter matrix which is computationally superior to the optimal projection method with comparable classification accuracy with respect OPSRC. The performance of the proposed approach is benchmarked with various publicly available face database.

Evolution of electronic structure with dimensionality in divalent nickelates

We investigate the evolution of electronic structure with dimensionality (d) of Ni-O-Ni connectivity in divalent nickelates, NiO (3-d), La2NiO4, Pr2NiO4 (2-d), Y2BaNiO5 (1-d) and Lu2BaNi5 (0-d), by analyzing the valence band and the Ni 2p core-level photoemission spectra in conjunction with detailed many-body calculations including full multiplet interactions. Experimental results exhibit a reduction in the intensity of correlation-induced satellite features with decreasing dimensionality. The calculations based on the cluster model, but evaluating both Ni 3d and O 2p related photoemission processes on the same footing, provide a consistent description of both valence-band and core-level spectra in terms of various interaction strengths. While the correlation-induced satellite features in NiO is dominated by poorly screened d(8) states as described in the existing literature, we find that the satellite features in the nickelates with lower dimensional Ni-O-Ni connectivity are in fact dominated by the over-screened d(10)L(2) states. It is found that the changing electronic structure with the dimensionality is primarily driven by two factors: (i) a suppression of the nonlocal contribution to screening; and (ii) a systematic decrease of the charge-transfer energy Delta driven by changes in the Madelung potential. [S0163-1829(99)09619-8].

The effect of structural dimensionality on the electrocatalytic properties of the nickel selenide phase

Nickel selenide (NiSe) nanostructures possessing different morphologies of wires, spheres and hexagons are synthesized by varying the selenium precursors, selenourea, selenium dioxide (SeO2) and potassium selenocyanate (KSeCN), respectively, and are characterized using X-ray diffraction, X-ray photoelectron spectroscopy, Raman spectroscopy, transmission electron microscopy and scanning electron microscopy techniques. Electrical measurements of a single nanowire and a hexagon carried out on devices fabricated by the focused ion beam (FIB) technique depict the semiconducting nature of NiSe and its ability to act as a visible light photodetector. The three different morphologies are used as catalysts for hydrogen evolution (HER), oxygen reduction (ORR) and glucose oxidation reactions. The wire morphology is found to be better than that of spheres and hexagons for all the reactions. Among the reactions studied, NiSe is found to be good for HER and glucose oxidation while ORR seems to terminate at the peroxide stage.

High Oxygen Storage Capacity and High Rates of CO Oxidation and NO Reduction Catalytic Properties of Ce1-xSnxO2 and Ce0.78Sn0.2Pd0.02O2-delta

Ce1-xSnxO2 (x = 0.1-0.5) solid solution and its Pd substituted analogue have been prepared by a single step solution combustion method using tin oxalate precursor. The compounds were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), and H-2/temperature programmed redution (TPR) studies. The cubic fluorite structure remained intact up to 50% of Sri substitution in CeO2, and the compounds were stable up to 700 C. Oxygen storage capacity of Ce1-xSnxO2 was found to be much higher than that of Ce1-xZrxO2 due to accessible Ce4+/Ce3+ and Sn4+/Sn2+ redox couples at temperatures between 200 and 400 C. Pd 21 ions in Ce0.78Sn0.2Pd0.02O2-delta are highly ionic, and the lattice oxygen of this catalyst is highly labile, leading to low temperature CO to CO2 conversion. The rate of CO oxidation was 2 mu mol g(-1) s(-1) at 50 degrees C. NO reduction by CO with 70% N-2 selectivity was observed at similar to 200 degrees C and 100% N-2 selectivity below 260 degrees C with 1000-5000 ppm NO. Thus, Pd2+ ion substituted Ce1-xSnxO2 is a superior catalyst compared to Pd2+ ions in CeO2, Ce1-xZrxO2, and Ce1-xTixO2 for low temperature exhaust applications due to the involvement of the Sn2+/Sn4+ redox couple along with Pd2+/Pd-0 and Ce4+/Ce3+ couples.

Studies in metal-ammonia reduction-5: Reduction and reductive methylation of some naphthoic acids

Birch reductio and reductive methylations of some substituted naphtholic acids have been examined. The factors influencing the mechanism of reduction process have been discussed. Some of the reduced naphthoic acids are useful synthons for synthesis.

Calciothermic reduction of TiO2: A diagrammatic assessment of the thermodynamic limit of deoxidation

Calciothermic reduction of TiO2 provides a potentially low-cost route to titanium production. Presented in this article is a suitably designed diagram, useful for assessing the degree of reduction of TiO2 and residual oxygen contamination in metal as a function of reduction temperature and other process parameters. The oxygen chemical potential diagram à la Ellingham-Richardson-Jeffes is useful for visualization of the thermodynamics of reduction reactions at high temperatures. Although traditionally the diagram depicts oxygen potentials corresponding to the oxidation of different metals to their corresponding oxides or of lower oxides to higher oxides, oxygen potentials associated with solution phases at constant composition can be readily superimposed. The usefulness of the diagram for an insightful analysis of calciothermic reduction, either direct or through an electrochemical process, is discussed. Identified are possible process variations, modeling and optimization strategies.

Electrochemical Reduction of Oriented Graphene Oxide Films: An in Situ Raman Spectroelectrochemical Study

Graphene oxide (GO) is assembled on a gold substrate by a layer-by-layer technique using a self-assembled cystamine monolayer. The negatively charged GO platelets are attached to the positively charged cystamine monolayer through electrostatic interactions. Subsequently, it is shown that the GO can be reduced electrochemically using applied DC bias by scanning the potential from 0 to -1 V vs a saturated calomel electrode in an aqueous electrolyte. The GO and reduced graphene oxide (RGO) are characterized by Raman spectroscopy and atomic force microscopy (AFM). A clear shift of the G band from 1610 cm-1 of GO to 1585 cm-1 of RGO is observed. The electrochemical reduction is followed in situ by micro Raman spectroscopy by carrying out Raman spectroscopic studies during the application of DC bias. The GO and RGO films have been characterized by conductive AFM that shows an increase in the current flow by at least 3 orders of magnitude after reduction. The electrochemical method of reducing GO may open up another way of controlling the reduction of GO and the extent of reduction to obtain highly conducting graphene on electrode materials.

Reduction of Permittivity in Epoxy Nanocomposites at Low Nano-filler Loadings

Experimental studies reveal a reduction in the values of permittivity for epoxy nanocomposites; at low filler loadings as compared to neat epoxy over a wide frequency range. This permittivity reduction is attributed to the interaction dynamics between nanoparticles: and epoxy chains at the interface region and interestingly, this interaction has also been found to influence the glass transition temperatures (T-g) of the examined nanocomposite systems. Accordingly, a dual nanolayer interface model for an epoxy based nanocomposite system is analyzed to explain the obtained permittivity characteristics.

Evidence for H2O2 as the product of reduction of oxygen by alternative oxidase in mitochondria from potato tubers

Oxygen Consumption by alternative oxidase (AOX), present in mitochondria of many angiosperms, is known to be cyanide-resistant in contrast to cytochrome oxidase. Its activity in potato tuber (Solarium tuberosum L.) was induced following chilling treatment at 4 degrees C.About half of the total O-2 consumption of succinate oxidation in such mitochondria was found to be sensitive to SHAM, a known inhibitor of AOX activity. Addition of catalase to the reaction mixture of AOX during the reaction decreased the rate of SHAM-sensitive oxygen consumption by nearly half, and addition at the end of the reaction released nearly half of the consumed oxygen by AOX, both typical of catalase action on H2O2. These findings with catalase suggest that the product of reduction of AOX is H2O2 and not H2O, as previously Surmised. In potatoes Subjected to chill stress (4 degrees C) for periods of 3, 5 and >= 8 days the activity of AOX in mitochondria increased progressively with a corresponding increase in the AOX protein detected by immunoblot of the protein.

NOx Reduction and Desorption Studies Under Electric Discharge Plasma Using a Simulated Gas Mixture: A Case Study on the Effect of Corona Electrodes

In this study, reduction and desorption of oxides of nitrogen (NOx) were conducted using an electrical discharge plasma technique. The study was carried out using a simulated gas mixture to explore the possibility of re-generation of used adsorbents by a nonthermal plasma desorption technique. Three different types of corona electrodes, namely, pipe, helical wire, and straight wire, were used for analyzing their effectiveness in NOx reduction/desorption. The pipe-type corona electrode exhibited a nitric oxide (NO) conversion of 50%, which is 1.5 times that of the straight-wire-type electrode at an energy density of 175J/L. The helical-wire-type corona electrode exhibited a NOx desorption efficiency almost 4 times that of the pipe-type electrode,indicating the possibility that corona-generated species play a crucial role in desorption.