A unifying criterion for unsupervised clustering and feature selection

Exploratory data analysis methods are essential for getting insight into data. Identifying the most important variables and detecting quasi-homogenous groups of data are problems of interest in this context. Solving such problems is a difficult task, mainly due to the unsupervised nature of the underlying learning process. Unsupervised feature selection and unsupervised clustering can be successfully approached as optimization problems by means of global optimization heuristics if an appropriate objective function is considered. This paper introduces an objective function capable of efficiently guiding the search for significant features and simultaneously for the respective optimal partitions. Experiments conducted on complex synthetic data suggest that the function we propose is unbiased with respect to both the number of clusters and the number of features.     

Effect of Powder Recycling on Defect Formation in Electron Beam Melted Alloy 718

Metallurgical and Materials Transactions A - The extent to which powder recycling can be permitted before risking a loss in performance of critical components is a major aspect for the viability of...     

Epoxy–lignin polyblends: Correlation between polymer interaction and curing temperature

A bisphenol A–polyamine hardener based epoxy adhesive (EP) was modified by polyblending with Kraft lignin (L). EP–L polyblends with an L content up to 40% by weight were cured at room temperature or above their glass transition temperature (T_g). Previous data have shown that the thermal and viscoelastic properties, as well as adhesive performance of the EP–L polyblends, are influenced by the curing temperature and by the L content in thermally cured polyblends. A reasonable explanation for the different behavior of EP–L polyblends as function of the curing temperature and their L content could be the enhanced degree of bonding between L and the EP network taking place at elevated temperature. This bonding was specifically considered to take place between L and possible unreacted amine groups of the hardener. Characterization of the EP–L polyblends was performed to search for evidence of irreversible chemical bonding between L and the EP network in thermally cured EP–L polyblends. FTIR studies, L extractibility from the crosslinked, polyblends, and quantitative data concerning the reactivity of L toward the polyamine hardener are discussed.     

Antibacterial activity of silver-modified natural clinoptilolite

The aim of the present work was to estimate the bactericidal activity and efficacy of silver pre-treated clinoptilolite-rich tuff from Marsid (Romania) in solid media (agar plates) against Gram-negative Escherichia coli ATCC 25922 and Gram-positive Staphylococcus aureus ATCC 25923. Two samples of natural clinoptilolite-rich tuff was first pre-treated with oxalic acid and sodium hydroxide solutions, respectively. The sample treated with oxalic acid was then exchanged with sodium chloride solution to obtain sodium form. Finally, both samples were exchanged with silver nitrate solution at room temperature for 24 h to obtain silver forms (P1-Ag⁺ and P2-Ag⁺) of clinoptilolite. The structure, morphology, and elemental composition of the pre-treated clinoptilolite samples were characterized by XRD, infrared (ATR-IR), SEM, and EDX analysis. The antibacterial activity was investigated by exposing E. coli and S. aureus in nutritive agar to the silver-clinoptilolite samples. Microorganisms were completely inhibited at 2 mg Ag⁺-clinoptilolite/mL nutritiv medium after 24 h of incubation at 37 °C. The silver-clinoptilolite sample derived from natural clinoptilolite pre-treated with oxalic acid (P1-Ag⁺) exhibit a stronger antibacterial effect in the presence of E. coli and the sample derived from natural clinoptilolite pre-treated with sodium hydroxide (P2-Ag⁺) in the presence of S. aureus.     

The Role of Matrix Metalloproteinases (MMP-8, MMP-9, MMP-13) in Periodontal and Peri-Implant Pathological Processes

Severe periodontitis, a destructive inflammatory disease of the supporting tissues of the teeth, ranks sixth in terms of global spread, affecting about 11% of the population. Metalloproteinases (MMPs) are extracellular matrix (ECM) macromolecules that are important in cellular development and morphogenesis, and they are capable of activating growth factors in their proximity, cell surface receptors, and adhesion molecules. MMPs are part of a major family of zinc-dependent endopeptidases, and their activity is modulated and regulated by certain inhibitors known as tissue metalloproteinase inhibitors (TIMPs). Because type I collagen is the major component of the periodontal extracellular matrix, special attention has been paid to the role of collagenases, especially MMP-8 and MMP-13 and gelatinases, MMP-2 and MMP-9, in periodontal diseases. In fact, MMP-8 (or collagenase 2) is currently one of the most promising biomarkers for periodontitis in oral fluids. Among them, salivary MMP-9 has been shown to be a more sensitive marker for periodontal inflammation during orthodontic treatment, which opens new perspectives in reducing periodontal hazards during such treatments. Both MMP-8 and MMP-9 are extremely valuable diagnostic tools in treating periodontitis, and future studies and healthcare policies should focus on implementing more accessible methods of chairside testing in order to reduce the prevalence of this disease.     

Host mRNA Analysis of Periodontal Disease Patients Positive for Porphyromonas gingivalis,Aggregatibacter actinomycetemcomitans and Tannerella forsythia

Periodontal disease is a frequent pathology worldwide, with a constantly increasing prevalence. For the optimal management of periodontal disease, there is a need to take advantage of actual technology to understand the bacterial etiology correlated with the pathogenic mechanisms, risk factors and treatment protocols. We analyzed the scientific literature published in the last 5 years regarding the recent applications of mRNA analysis in periodontal disease for the main known bacterial species considered to be the etiological agents: Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans and Tannerella forsythia. We identified new pathogenic mechanisms, therapeutic target genes and possible pathways to prevent periodontal disease. The mRNA analysis, as well as the important technological progress in recent years, supports its implementation in the routine management of periodontal disease patients.     

Cyclic organosilicon compounds

Cyclic organosilicon compounds were obtained by the coupling reaction of 1,4-dilithium-1,1,4,4-tetraphenylbutane and organochlorosilanes with general formula RR SiCl_z (R=methyl, R=ethyl, phenyl, vinyl, p-tolyl). The chemical analysis, molecular weight determination, IR spectra have shown that new compounds have been formed by ring closure through a siloxane bond. The thermogravimetric analysis indicates a thermal stability of compounds up to 360°C.     

Damaging Prediction of Difficult-to-Work Aluminum Alloys During Equal Channel Angular Pressing

Severe plastic deformation (SPD) is a well-established method in the recent years for grain refinement in metallic materials. Equal channel angular pressing (ECAP) is one of the most effective SPD techniques. Inherent failures of ECAP, consisting in billet damage, take place if not made a correct process design. In this article, the evolution of damaging for a difficult-to-work Al-Mg alloy during ECAP was investigated. A tridimensional finite element analysis was performed for four different die designs to study the influence of die geometry and process parameters on billet damaging. To validate modeling we used, load level and strain distribution were depicted. Experimental tests were performed to test the numerical prediction. The results show that cracking may be reduced or eliminated by inner fillet corner of the die channels. It was demonstrated that the predicted results were in good agreement with experimental data obtained for 5052 aluminum alloy. The direct effect of knowledge about load and damaging during ECAP is to prevent both tool and billet damage.     

Developing Web Databases for Aboriginal Language Preservation

This article discusses the development of integrated multilingualWeb databases to help the preservation of the Native Americanlanguage East Cree. The creation of digital online resourcesfor threatened aboriginal languages presents many technical,educational and ethical challenges. We focus here on the technicalchallenges in order to discuss both the problems encounteredin this particular context, and the solutions we have consideredand explored. We illustrate our discussion with examples froman Oral Stories Database we developed in collaboration withCree education consultants and speakers in 2002–04. Weadvocate an approach that includes fast-prototyping, open-sourcedevelopment, and design for the database engine that balancesspeed, availability, features, and resources. We discuss theimpact the combination of this technical approach and the participatoryaction research method is having on language maintenance.