Boosting support vector machines for imbalanced data sets

Real world data mining applications must address the issue of learning from imbalanced data sets. The problem occurs when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed vector spaces or lack of information. Common approaches for dealing with the class imbalance problem involve modifying the data distribution or modifying the classifier. In this work, we choose to use a combination of both approaches. We use support vector machines with soft margins as the base classifier to solve the skewed vector spaces problem. We then counter the excessive bias introduced by this approach with a boosting algorithm. We found that this ensemble of SVMs makes an impressive improvement in prediction performance, not only for the majority class, but also for the minority class.     

Tetrahydroindoles as Multipurpose Screening Compounds and Novel Sirtuin Inhibitors

Indoles are privileged structures in medicinal and bioorganic chemistry that are particularly well suited to serve as platforms for diversity. Among many other therapeutic areas, the indole scaffold has been used to design aromatic compounds useful to interfere with enzymes engaged in the regulation of substrate acylation status, such as sirtuins. However, the planarity of the indole ring is not necessarily optimal for all target enzymes, especially when functionalization with aromatic side chains is required. Replacement of flat scaffolds by nonplanar molecular cores dominated by sp³ hybridization is a common strategy to avoid the disadvantages associated with poor solubility and high promiscuity, while covering less-well-explored areas of chemical space. Thus, we synthesized fragment-like tetrahydroindoles suitable for fragment-based drug discovery as well as a well-characterized small library intended as multipurpose screening compounds. For proof of principle, these compounds were screened against sirtuins 1–3, enzymes known to be addressable by indoles. We found that 2,6,6-trimethyl-4-oxo-4,5,6,7-tetrahydro-1H-indole-3-carboxamides are potent and selective SIRT2 inhibitors. Compound 16 t displayed an IC₅₀ value of 0.98 μm and could serve as exquisite starting point for hit-to-lead profiling.     

Near-Field Electron Energy Loss Spectroscopy in Porous Silicon

First results of an Electron Energy Loss Spectroscopy in the Near Field (NFEELS) mode of n⁺ porous silicon are described here. Sequences of EELS spectra in the low loss energy range (0–30 eV) were recorded, using a scanning transmission electron microscope, as the e-beam was scanned across a nano-hole surrounded by Si platelets. This technique is shown to be very sensitive to spectral and spatial changes in the electromagnetic field distribution outside the surface of nanoparticles, governed by their local nature and shape.     

Variable energy positron beam study of Xe-implanted uranium oxide

Doppler broadening of annihilation gamma-line combined with a slow positron beam was used to measure the momentum density distribution of annihilating pair in a set of sintered UO₂ samples. The influence of surface polishing, of implantation with 800-keV ¹³⁶Xe²⁺ at fluences of 1 × 10¹⁵ and 1 × 10¹⁶ Xe cm^?2, and of annealing were studied by following the changes of the momentum distribution shape by means of S and W parameters. The program used for this purpose was VEPFIT. At the two fluences in the stoichiometric as-implanted UO₂, formation of Xe bubbles was not detected. The post-implantation annealing and over-stoichiometry in the as-implanted sample caused Xe precipitation and formation of Xe bubbles.     

Detection of clenbuterol in beef meat,liver and kidney by mid‐infrared spectroscopy (FT‐Mid IR) and multivariate analysis

Mid‐infrared spectroscopy (FT‐Mid IR) coupled with multivariate analysis was used to predict clenbuterol in beef meat, liver and kidney. A SIMCA model was also developed to discriminate between pure (beef meat, liver and kidney) and spiked with clenbuterol samples (beef meat‐clenbuterol, liver‐clenbuterol and kidney‐clenbuterol). The best models to predict clenbuterol concentrations were obtained using the partial least squares algorithm (PLS) with a R² > 0.9 and SEC and standard error of prediction <0.296 and 0.324, respectively. The SIMCA model used to discriminate pure and spiked with clenbuterol samples showed 100% correct classification rate. Methods detection limit was 2 μg kg^?1. FT‐Mid IR coupled with chemometrics could be a simple and rapid screening tool for monitoring clenbuterol in beef meat, liver and kidney implicated in food poisoning. This method could be use for screening purposes.     

Sharing the ‘fame’ of ISO standard adoption: quality supply chain effects evidence

The objective of this paper is to provide an empirical evidence for whether the hierarchical position (determined by firm’s quality strength) of different types of quality supply chain categories implies the same hierarchy in terms of economic gains. Using data collected in France through a survey, we distinguish between four types of quality supply chain categories that help us to empirically construct a relationship between firms and their suppliers (quality certified/non-quality certified). Our findings, from a two-regime switching model, reveal that there is a positive correlation between the hierarchical positions of the quality supply chain categories and the impact on firm performance (measured by profit per employee).     

Synthesis of calcium hydroxyapatite from calcium carbonate and different orthophosphate sources: A comparative study

The synthesis of calcium hydroxyapatite (Ca-HA) starting from calcium carbonate and different orthophosphate sources, including orthophosphoric acid, potassium, sodium and ammonium dihydrogen orthophosphates, was investigated under ambient conditions. The reaction started with calcium carbonate dissolution in an acid medium, followed by rapid precipitation of calcium cations with orthophosphate species to form calcium phosphate based particles which were in the size range of 0.4–1 μm. These particles then agglomerated into much larger ones, up to 350 μm in diameter (aggregates). These aggregates possessed an unstable porous structure which was responsible for the porosity of the final products. The highest specific surface area and pore volume were obtained with potassium dihydrogen orthophosphate. On the other hand, orthophosphoric acid led to the highest dissolution of calcium carbonate and the complete precipitation of orthophosphate species. Under ambient conditions, calcium phosphate based solid products of low crystallinity were formed. Different intermediates were identified and a reaction pathway proposed.     

Sensory profiles and preference analysis in ornamental horticulture: The case of the rosebush

The context of ornamental horticulture is considered in order to extend the techniques of sensory and preference evaluation by taking the rosebush as a plant model. In a preliminary study (Boumaza, Demotes-Mainard, Huché-Thélier, & Guérin, 2009), a sensory evaluation was conducted in order to set up a list of attributes. Subsequently, this list was adapted to assess 10 rosebushes. After the control of the panel performance using a multivariate strategy of analysis, the average scores were used in product mapping. The evaluation of the preferences with regard to these rosebushes was undertaken: 253 subjects were asked to rank the products by decreasing order of liking. Thereafter, the preference data were subjected to an internal preference mapping and a cluster analysis. Six homogeneous segments of consumers were eventually retained. By way of performing an external preference mapping, the average ranks were regressed upon the sensory attributes using principal component regression: the preferences of 67% of the consumers were satisfactorily explained by the attributes.