A self-adaptive synthetic over-sampling technique for imbalanced classification

Traditionally, in supervised machine learning, (a significant) part of the available data (usually 50%-80%) is used for training and the rest—for validation. In many problems, however, the data are highly imbalanced in regard to different classes or does not have good coverage of the feasible data space which, in turn, creates problems in validation and usage phase. In this paper, we propose a technique for synthesizing feasible and likely data to help balance the classes as well as to boost the performance in terms of confusion matrix as well as overall. The idea, in a nutshell, is to synthesize data samples in close vicinity to the actual data samples specifically for the less represented (minority) classes. This has also implications to the so-called fairness of machine learning. In this paper, we propose a specific method for synthesizing data in a way to balance the classes and boost the performance, especially of the minority classes. It is generic and can be applied to different base algorithms, for example, support vector machines, k-nearest neighbour classifiers deep neural, rule-based classifiers, decision trees, and so forth. The results demonstrated that (a) a significantly more balanced (and fair) classification results can be achieved and (b) that the overall performance as well as the performance per class measured by confusion matrix can be boosted. In addition, this approach can be very valuable for the cases when the number of actual available labelled data is small which itself is one of the problems of the contemporary machine learning.     

Crystallization of progesterone polymorphs using polymer-induced heteronucleation (PIHn) method

Progesterone is a natural hormone steroid used in humans for several treatments and in livestock for artificial insemination, which exhibits two polymorphic forms at ambient conditions: form 1 and form 2. Form 2 is metastable and more soluble than form 1; however, it is not suitable to use as powder raw material because it transforms into form 1 by the effects of grinding. A polymorphic screening of progesterone based on polymer-induced heteronucleation method was performed as an alternative to prepare the metastable form. Polyvinyl alcohol, hydroxypropyl methylcellulose (HPMC), dextran, gelatin, polyisoprene (PI) and acrylonitrile-butadiene (NBR) copolymer were used. Crystals were prepared from 0.5, 10 and 40?mg/mL solutions in acetone at room temperature by solvent evaporation. The samples were characterized by X-ray powder diffraction, differential scanning calorimetry (DSC), scanning electron microcopy and attenuated total reflectance infrared Fourier transform spectroscopy. Form 1 was nucleated from 40?mg/mL solutions on the six polymers and from 10?mg/mL solutions on PI and NBR. The mixture of form 1 and form 2 was obtained from 10?mg/mL solution on HPMC, dextran and gelatin and from 0.5?mg/mL solution crystallizations. Therefore, the polymeric devices, which crystallized the metastable and more soluble polymorph (2) of progesterone, would be a promissory alternative for the pharmaceutical applications.     

Enzyme activity and partial characterization of proteases obtained from Bromelia karatas fruit and compared with Bromelia pinguin proteases

Food Science and Biotechnology - The enzymatic activity and partial characterization of proteases from Bromelia karatas fruits were evaluated and compared with Bromelia pinguin proteases. The...     

A new mobile robot control approach via fusion of control signals.

This paper proposes an alternative approach to address the problem of coordinating behaviors in mobile robot navigation: fusion of control signals. Such approach is based on a set of two decentralized information filters, which accomplish the data fusion involved. Besides these two fusion engines, control architectures designed according to this approach also embed a set of different controllers that generate reference signals for the robot linear and angular speeds. Such signals are delivered to the two decentralized information filters, which estimate suitable overall reference signals for the robot linear and angular speeds, respectively. Thus, the background for designing such control architectures is provided by the nonlinear systems theory, which makes this approach different from any other yet proposed. This background also allows checking control architectures designed according to the proposed approach for stability. Such analysis is carried out in the paper, and shows that the robot always reaches its final destination, in spite of either obstacles along its path or the environment layout. As an example, a control architecture is designed to guide a mobile robot in an experiment, whose results allows checking the good performance of the control architecture and validating the design approach proposed as well.     

Feedback tuning of bifurcations

The present paper studies a feedback regulation problem that arises in at least two different biological applications. The feedback regulation problem under consideration may be interpreted as an adaptive control problem for tuning bifurcation parameters, and it has not been studied in the control literature. The goal of the paper is to formulate this problem and to present some preliminary results.     

Kinetic model for water/oil absorption of mesquite gum (Prosopis juliflora) and gum arabic (Acacia senegal)

The water and oil uptake of mesquite and arabic gums in powdered form was studied at temperatures of 23, 35 and 45°C. A previously proposed equation to predict osmotic equilibrium was tested using the experimental data with both gums and a good statistical fit was obtained. Mesquite gum showed the highest water and oil absorption at all temperatures studied. Temperature dependence of the reciprocal of the S₁ and WL_∞ were determined using an Arrhenius equation. The activation energy for water and oil absorption for gum arabic was 21.98 and 39.57 kJ mol⁻¹, compared to that of mesquite gum having values of 15.79 and 46.16 kJ mol⁻¹, respectively. A second order kinetic model was obtained for water and oil absorption for both gums.     

Extrusion-rotomolding of plasticized PVC formulations

In this work, we present a novel process to produce plasticized PVC articles. Plasticized PVC dispersion resin (dry blend) was extruded and poured into a hot mold. The mold was transferred to a rotomolding machine, where the final geometry was obtained. Formulations using commercial PVC suspension resin and three different types of plasticizers were prepared under different processing conditions. The physical appearance of the articles and the tensile properties of the materials were evaluated.     

A typology of propagation of changes on the structure of a multiregional economic system: the case of the European Union, 1975–1985

The effects of structural change in all sectors of some regions in multiregional input-output systems are explored in this paper. For each region, the changes in direct inputs are decomposed into a matrix of intraregional changes and a matrix of interregional backward linkage changes. This distinction provides for a decomposition of changes in the Leontief inverse into a sum of matrices representing the matrices of intraregional and interregional fields of influence of changes. In turn, this process facilitates the identification of changes in gross outputs that may be traced to changes in intra- and interregional structural changes. The analysis is applied to the European input-output tables for 1975 and 1985.This paper was presented at the 1996 annual meeting of the Western Regional Science Association     

Spatially resolved infrared spectroscopy: a novel technique for in situ study of spatial surface coverage during CO oxidation on supported catalysts

A spatially resolved infrared (IR) imaging technique to monitor the linear adsorbed CO coverage on supported catalyst surface combining an IR bandpass filter and an IR thermography camera has been developed. Images acquired during the CO adsorption/desorption and ignition indicate that the technique provides an excellent method to image the change of surface coverage with a spatial resolution. It is expected that the combination of infrared thermography with spatially resolved imaging of surface coverage will provide a deeper insight in the dynamics of spatio-temporal patterns on heterogeneous catalysts.