Influence diagnostics in multiple discriminant analysis

Various influence diagnostics in Multiple Discriminant Analysis can be found in the literature. Almost all of them are based on the overall probability of misclassification. Two case-deletion diagnostics which evaluate the effect of the omission on the linear functions which determine Fisher’s Linear Discriminant Rule are proposed in this paper. Both measures are based on the L ₂-norm: The first diagnostic is calculated from the data set and the second diagnostic is calculated in the minimum hypercube which covers it.     

Study of the hydrolysis and condensation of γ-Aminopropyltriethoxysilane by FT-IR spectroscopy

The hydrolysis and condensation reactions of γ-APS have been studied in different acid content aqueous solution by using Fourier Transform infrared (FT-IR) spectroscopy. The hydrolysis of γ-APS under the studied conditions can be followed by the increase of the ethanol band located at 882 cm⁻¹ and the decrease of the band due to the ρ(CH₃) of γ-APS molecules located at 959 cm⁻¹. Hydrolysis reaction is faster by increasing both H₂O and acid concentrations, and it is completed when 3 moles of H₂O per mole of γ-APS are used. The increase of the vibrational band located at 1146 cm⁻¹ shows that condensation of the hydrolysed γ-APS molecules take place forming linear chains in poorly cross-linked structures. Besides, both 8-membered cyclic siloxane formations and poorly cross-linked structures are formed and increase as the water and acid content are increased. On the other hand, highly connected cross-linked structures do not appear due to the steric hindrance of the non-hydrolysable aminopropyl group. The silanol band shows that hydrolysis is faster than condensation except for samples with the lowest H₂O content.     

Crystallization study of chemically vapour-deposited amorphous silicon films by in situ x-ray diffraction

The solid phase crystallization kinetics of chemically vapour-deposited amorphous silicon films were studied by in situ X-ray diffraction. We determined the crystalline volume directly from the Bragg peak intensities at various times during isothermal annealing in the temperature range 578 °C < T < 658 °C. From these experiments we deduced that the crystallization was due to nucleation predominantly at the substrate-film interface followed by crystal growth perpendicular to this interface. The crystal growth rate was thermally activated with an activation energy E_v of 3.1 eV. A strong 〈111〉 preferred orientation of the growing polycrystal was observed and the grain size remained constant at about 60 nm. Evidence of stresses at the amorphous-crystalline interface during the early stages of crystallization was observed. A comparison with previous conductivity measurements is also carried out.     

Diffusion bonding of an aluminium-lithium alloy (AA8090) using aluminium-copper alloy interlayers

Diffusion bonds have been produced between sheets of an Al-Li-Cu-Mg-Zr alloy using aluminium-4% copper vapour deposited metallic interlayers. Microstructural changes occurred both in the parent alloy and in the bond interface after diffusion bonding cycles and post-bonding heat treatments were analysed. Different metallographic techniques (light microscopy, scanning and transmission electron microscopy) have been used. Diffusion bonding trials were carried out using the same alloy (AA8090), both in non-superplastic (T6) and superplastic conditions. Differences in their behaviours in relation to diffusion bonding were observed.     

Learning Recursive Bayesian Multinets for Data Clustering by Means of Constructive Induction

This paper introduces and evaluates a new class of knowledge model, the recursive Bayesian multinet (RBMN), which encodes the joint probability distribution of a given database. RBMNs extend Bayesian networks (BNs) as well as partitional clustering systems. Briefly, a RBMN is a decision tree with component BNs at the leaves. A RBMN is learnt using a greedy, heuristic approach akin to that used by many supervised decision tree learners, but where BNs are learnt at leaves using constructive induction. A key idea is to treat expected data as real data. This allows us to complete the database and to take advantage of a closed form for the marginal likelihood of the expected complete data that factorizes into separate marginal likelihoods for each family (a node and its parents). Our approach is evaluated on synthetic and real-world databases.     

Increasing the Complexity of Magnetic Core/Shell Structured Nanocomposites for Biological Applications

This Review article ponders core/shell structured nanoparticles that can be prepared with features that combine properties of different materials, including ligands that enhance their biocompatibility. These nanocomposites are not classified in terms of synthesis, but rather by how these features are distributed in the final morphology, attending to connected or isolated materials that end up in interacting or not‐interacting functionalities. In particular, we have focused on magnetic core/shell‐structured particles with a directly connected, coupled, or isolated second functionality. The current progress on methods in colloidal solution that have allowed the great development of these multifunctional magnetic and active spheres on biological and biomedical fields is reported.     

Scientific and technological performance evaluation of the Spanish Council for Scientific Research (CSIC) in the field of Biotechnology

An evaluation of the Spanish CSIC performance in Biotechnology, as compared with those of the French CNRS and the Italian CNR, has been carried out to determine the balance between the generation of scientific knowledge and the transfer of technology. This study shows a high scientific productivity mostly in journals with moderate impact factor, a low generation of patents and an insufficient transfer of knowledge to the Spanish companies. Other indicators confirm the existence of competitive human resources in biotechnological research producing scientific knowledge of interest for the development of patents and that cooperates successfully at European level.