Combination Of Ensemble Research Articles

Forecasting the NN5 time series with hybrid models

Abstract We propose a simple way of predicting time series with recurring seasonal periods. Missing values of the time series are estimated and interpolated in a preprocessing step. We combine several forecasting methods by taking the weighted mean of forecasts that were generated with time-domain models which were validated on left-out parts of the time series. The hybrid model is a combination of a neural network ensemble, an ensemble of nearest trajectory models and a model for the 7-day cycle. We apply this approach to the NN5 time series competition data set.

From an aggregate to a brand network: a study of the brand portfolio at L'Oréal

Our purpose is to contribute to the understanding of brand-portfolio management by examining the brand-portfolio strategies of a world-leading company. We started to work on a case study with L'Oréal. Our research focused on two questions: (1) what reasons lead L'Oréal to develop a brand-portfolio strategy?; (2) how can brand-portfolio management create a higher and stronger level of competitive advantage for this company? The results show that an aggregation of brands is not in itself a brand portfolio. The juxtaposition of brands is one of, but not the sole, element necessary for the development of a brand portfolio, which is a combination of a brand ensemble and key competences born out of organisational savoir-faire. By analysing the evolution of the L'Oréal brand portfolio, this work shows how the brand combination within a portfolio is a key factor for company development, growth, and risk management. It is a crucial phase to understand L'Oréal's competitive advantage.

Photophysics of the Red Chromophore of HcRed: Evidence for Cis−Trans Isomerization and Protonation-State Changes

HcRed is a dimeric intrinsically fluorescent protein with origins in the sea anemone Heteractis crispa. This protein exhibits deep red absorption and emission properties. Using a combination of ensemble and single molecule methods and by varying environmental parameters such as temperature and pH, we found spectroscopic evidence for the presence of two ground state conformers, trans and cis chromophores that are in thermal equilibrium and that follow different excited-state pathways upon exposure to light. The photocycle of HcRed appears to be a combination of both kindling proteins and bright emitting GFP/GFP-like proteins: the trans chromophore undergoes light driven isomerization followed by radiative relaxation with a fluorescence lifetime of 0.5 ns. The cis chromophore exhibits a photocycle similar to bright GFPs and GFP-like proteins such as enhanced GFP, enhanced YFP or DsRed, with radiative relaxation with a fluorescence lifetime of 1.5 ns, singlet-triplet deactivation on a microsecond time scale and solvent controlled protonation/deprotonation in tens of microseconds. Using single molecule spectroscopy, we identify trans and cis conformers at the level of individual moieties and show that it is possible that the two conformers can coexist in a single protein due to the dimeric nature of HcRed.

Histone Post-Translational Modifications Buried within the Nucleosome DNA-Histone Interface Facilitate Nucleosome Disassembly

A number of histone post-translational modifications are located in the nucleosome DNA-histone interface. These modifications include the acetylation of histone H3 at K115 and K122, which are located near the dyad symmetry axis. While these modifications are implicated in DNA repair, the molecular mechanisms by which these modifications are involved in DNA repair processes are not well-understood. We are investigating the hypothesis that these modifications biophysically alter the nucleosome to facilitate DNA repair. We introduced these modifications into histone H3 by protein ligation and incorporated them into single nucleosomes and nucleosome arrays. We employed a combination of ensemble and single molecule methods to determine that these modifications reduce DNA-histone binding free energy and enhance nucleosome disassembly induced by the mechanical unwrapping of nucleosomal DNA, or by the human mismatch repair complex hMSH2-hMSH6. These studies suggest that modifications in the nucleosome DNA-histone interface function to reduce DNA-histone interactions to facilitate nucleosome disassembly for DNA repair.This work was supported by a Career Award in the Basic Biomedical Sciences from the Burroughs Welcome Fund to MGP and by the NIH, R01-GM083055, to MGP and JJO.

Brands portfolios and competitive advantage: an empirical study

PurposeThe paper seeks to contribute to the understanding of brand portfolio management by examining the brand portfolio strategies of four leading cosmetics companies. The research focuses on two questions: what reasons lead companies to develop, or not, a brand portfolio strategy, and how brand portfolio management can create a higher and stronger level of competitive advantage that is harder to grasp and imitate.Design/methodology/approachThe paper utilises an exploratory approach by means of case studies. Data were collected from the following companies: L'Oréal, Clarins, Estée Lauder and Wella. The research involved in‐depth interviews with 33 company directors.FindingsThe research results show that an aggregation of brands is not in and of itself a brand portfolio. The juxtaposition of brands is one of the elements, but not the sole element, necessary for the development of a brand portfolio, which is a combination of a brand ensemble and key factors born out of organisational savoir‐faire.Research limitations/implicationsThe results validate the existence of a link between brand portfolios and competitive advantage, a link based on the existence of four key factors identified in the research.Practical implicationsA model is proposed to assist managers in better understanding and controlling brand regrouping, because the research illustrates the benefits for a company that executes well‐coordinated brand management.Originality/valueThis research fits into the complex context of strategic/marketing relationships and broadens the field of brand analysis, notably its strategic dimension. The contribution of this research is to show how a brand portfolio can create a stronger and higher level of competitive advantage, which is more difficult to copy.

Motor step size and ATP coupling efficiency of the dsDNA translocase EcoR124I

The Type I restriction-modification enzyme EcoR124I is an archetypical helicase-based dsDNA translocase that moves unidirectionally along the 3′–5′ strand of intact duplex DNA. Using a combination of ensemble and single-molecule measurements, we provide estimates of two physicochemical constants that are fundamental to a full description of motor protein activity—the ATP coupling efficiency (the number of ATP consumed per base pair) and the step size (the number of base pairs transported per motor step). Our data indicate that EcoR124I makes small steps along the DNA of 1 bp in length with 1 ATP consumed per step, but with some uncoupling of the ATPase and translocase cycles occurring so that the average number of ATP consumed per base pair slightly exceeds unity. Our observations form a framework for understanding energy coupling in a great many other motors that translocate along dsDNA rather than ssDNA.

Level statistics in disordered linear networks

We consider a linear electrical impedance network, and study its frequency of nearest level statistics, P(s). We find that this probability density evolves from a sum of gaussians at small disorder, to a linear combination of Poisson and gaussian orthogonal ensemble for large values of the disorder.

An Ensemble and Single-molecule Fluorescence Spectroscopy Investigation of Calcium Green 1, a Calcium-ion Sensor

The calcium-ion indicator dye, Calcium Green 1 (CG-1), has been characterized using a combination of ensemble and single-molecule optical spectroscopy measurements. In terms of ensemble measurements, CG-1 demonstrated a strong increase in fluorescence emission as a function of increasing [Ca(2+)]. This was accompanied by a change in the relative proportions of two chemical forms of the dye, each with a different fluorescence lifetime, which were found to co-exist in solution. From single-molecule fluorescence measurements, it was found that the fluorescence intensity and photobleaching time (on-time) of each CG-1 molecule was invariant with [Ca(2+)] and that changes in ensemble fluorescence intensity simply correlates with the number of fluorescent molecules in solution. These results are compared with that of the related system, Calcium Green 2 (CG-2), and the mechanisms of operation of these two indicator dyes are discussed.

Head-worn, trimodal device to increase transcript accuracy in a voice recognition system and to process unvocalized speech

A voice recognition device and method allows position-stabilized capture of spoken sounds with great repeatability and accuracy. The voice recognition device may additionally provide two channels of lip movement information to supplement the usual audible speech component recognition system in selecting the proper pairing of data input to text output. The voice recognition device may provide a further channel of information about the speech generating motions via an ultrasonic injection of sound into the vocal cavity and subsequent decoding of the emitted sound after injection. The ultrasonic injection and decoding may also used to provide audible clues as to the unvoiced sound formed by speaking when the vocal cords are not energized. The ensemble of electronic equipment upon the bail band may be in microcircuit form, including placing the components on a copper layer polyimide flexible strip. The side camera and “other side” illuminator LED may be on thin copper polyimide strips attached to the main electronics ensemble, and a set of thin polyimide conductors would conduct power into the ensemble and the signals out of the ensemble through one of the bail band ends, into the ear piece and down the connector to the associated computer equipment and may also supply the power for the electronic ensemble. The electronic ensemble may be potted with a thin layer of elastomer, such as translucent silicone, and provide a moisture barrier and physical protection for the ensemble, while still offering a very light visual weight to the combination of the electronic ensemble and the bail band.

The birth of a bubble: A molecular simulation study

We study the nucleation of a bubble in a metastable Lennard-Jones (LJ) fluid, confined to a spherical pore with wetting walls, by a combination of grand canonical, canonical ensemble, and gauge cell Monte Carlo simulation methods complemented by the Voronoi-Delaunay tessellation analysis of statistical geometry of intermolecular cavities. We construct the isotherm of confined fluid in the form of a continuous van der Waals' loop, in which the unstable backward trajectory between the spinodals corresponds to bubble states. We show that as the degree of metastability increases and the fluid becomes progressively stretched, the decrease of fluid density is associated with the evolution of a population of interstitial intermolecular cavities. At the spinodal, the fluid becomes mechanically unstable: Interstitial cavities partly coalesce into a larger cavity located due to the system symmetry around the pore center. This cavity represents a bubble embryo, which grows at the expense of interstitial cavities. The nucleation barrier is calculated by direct thermodynamic integration along the isotherm. We compare our simulation results to the predictions of the classical nucleation theory and experiments on capillary condensation-evaporation of nitrogen in pores of hybrid organic-inorganic mesoporous molecular sieve HMM-3.

Synthesis of and excited state processes in multichromophoric dendritic systems

Rigid polyphenylene dendrimers substituted with dyes are suitable substrates for the investigation of interchromophore photophysical processes as they contain discrete numbers of chromophores in well-defined spatial relationships to each other. In this contribution, we provide an overview of the synthesis of such materials and show how a combination of ensemble and single-molecule spectroscopic investigations can be used to elucidate the interchromophore excitation energy transfer processes in them.

GC analysis of human breath with a series-coupled column ensemble and a multibed sorption trap.

The combination of a tandem column ensemble and an on-line microsorption trap is used for the analysis of organic compounds in human breath samples. The four-bed sorption trap uses a series of discreet sorption beds containing three grades of graphitized carbon and a carbon molecular sieve to quantitatively remove most organic compounds from 0.8-L breath samples. The trap is then heated to 300 degrees C in approximately 1.5 s and maintained at this temperature for 10 s. The resulting vapor plug width is in the range 0.7-1.3 s for the compounds found in the breath samples. The separation is performed with a 15-m-long, 0.25-mm-i.d. capillary using a 0.5-microm-thick film of nonpolar dimethyl polysiloxane coupled in series to a polar column, either trifluoropropylmethyl polysiloxane or poly(ethylene glycol). Both column combinations are successful in separating the early-eluting compounds acetone, isoprene, pentane, methyl alcohol, and ethyl alcohol, which are all common in breath samples. The poly(ethylene glycol) combination gave better separation but showed relatively fast deterioration for repeated analysis of wet samples. Breath samples were obtained under different conditions (smoker, nonsmoker, gum chewer), and 25 compounds were identified in the various samples. Many additional peaks are observed but not identified. Analytical curves (log-log) of peak area versus sample volume for test compounds are linear in the range 80-800 cm3. Detection limits (3sigma) for several volatile compounds in 800-cm3 samples are in the 1-5 ppb range.

Target Immunity during Mu DNA Transposition: Transpososome Assembly and DNA Looping Enhance MuA-Mediated Disassembly of the MuB Target Complex

The Mu transpososome can distinguish between proximal and distal DNA during the selection of a site for transposition. This phenomenon, termed target immunity, involves MuA-stimulated removal of MuB oligomers from sites near the Mu genome. Using a combination of ensemble and single-molecule fluorescence methods, we show that the MuA tetramer can stably associate with the DNA-bound MuB oligomer and is more efficient than monomeric MuA at stimulating the dissociation of MuB from DNA. In addition, we demonstrate that DNA looping is essential for efficient disassembly of the MuB oligomer. We propose a model in which the MuA tetramer forms a multivalent complex with the MuB oligomer and catalyzes the processive removal of MuB from DNA.

A microstrip subarray for a DCS1800 base station

This paper presents the design and development of a four-element, corporate feed, linear subarray for possible use in a base-station antenna developed for the DCS1800 system. The subarray, with the main beam 30° off boresite toward the ground, uses strip-slot-foam inverted-patch construction to increase the operational bandwidth. The design was carried out using a combination of the commercial software packages Ensemble and PCAAD. Experimental and simulated results for return loss and radiation patterns are presented and compared; a good agreement is seen in the return loss results, and the main beam is tilted as predicted; however, the measured sidelobe levels are higher than desired. ©1999 John Wiley & Sons, Inc. Microwave Opt Technol Lett 20: 169–173, 1999.

Non-equilibrium molecular dynamics for simulating permeation of gas mixtures through nanoporous carbon membranes.

This paper deals with the computer modeling of permeation of a gas mixture through a nanoporous membrane by introducing a novel approach of non-equilibrium molecular dynamics (NEMD). A combination of the grand canonical ensemble (GCE or μVT) Monte Carlo (MC) technique and the boundary-driven NEMD method (Hafskjold et al., 1993; Ikeshoji and Hafskjold, 1994) is proposed for simulation of gas permeation based on the selective adsorption and surface diffusion mechanism

Promotion and poisoning of the reaction of methanol on clean and modified platinum (100)

The reactivity of Pt(100) towards methanol depends on surface structure and the nature and amount of coadsorbed reaction modifiers. We studied the adsorption and reaction of CH 3OH in ultrahigh vacuum on the clean hex and (1 × 1) surfaces of Pt(100) and with coadsorbed O, Bi, and CO with thermal desorption spectroscopy and high resolution electron energy loss spectroscopy. Methanol adsorbs molecularly on Pt(100) at 100 K and decomposes to adsorbed H and CO in the temperature range of 170–220 K. The vibrational spectra show evidence for surface interactions involving both the hydroxyl (O-H … Pt) and methyl (C-H … Pt) hydrogens of molecular methanol. Under static conditions (no desorption of reaction products) as much as 0.085 ± 0.01 monolayers of methanol react on the clean hex surface. Chemical saturation of the surface, which occurs when the total coverage of surface reaction products reaches about one-half of a monolayer, limits the maximum extent of reaction on the hex, (1 × 1), and O-covered surfaces. The (1 × 1) surface enhances methanol reaction by factors of 1–3.4 compared to reaction following the same exposure on the hex surface. Reaction with preadsorbed oxygen leads to a methoxy intermediate (CH 3O) and enhancement factors of 1–2.5 for oxygen coverages below 0.15 and above 0.3 monolayers. Coadsorbed oxygen enhances methanol reaction stoichiometrically, by scavenging H atoms to form OH and H 2O, and nonstoichiometrically, most likely by local lifting of the hex reconstruction. Oxygen coverages between 0.15 and 0.3 monolayers inhibit the reaction by factors of 1—0.85 through blocking of reactive sites. Both Bi and CO inhibit the reaction by factors of 1—0.05 through a combination of ensemble and site blocking effects. Methanol decomposition requires an ensemble of 5 adsorption sites, as determined by Bi coadsorption experiments. The adsorbate/substrate combinations in this study represent model electrochemical situations, and we briefly discuss the implications of these findings toward methanol electrocatalysis.