Improving accuracy and speed of optimum-path forest classifier using combination of disjoint training subsets

The Optimum-Path Forest (OPF) classifier is a recent and promising method for pattern recognition, with a fast training algorithm and good accuracy results. Therefore, the investigation of a combining method for this kind of classifier can be important for many applications. In this paper we report a fast method to combine OPF-based classifiers trained with disjoint training subsets. Given a fixed number of subsets, the algorithm chooses random samples, without replacement, from the original training set. Each subset accuracy is improved by a learning procedure. The final decision is given by majority vote. Experiments with simulated and real data sets showed that the proposed combining method is more efficient and effective than naive approach provided some conditions. It was also showed that OPF training step runs faster for a series of small subsets than for the whole training set. The combining scheme was also designed to support parallel or distributed processing, speeding up the procedure even more. © 2011 Springer-Verlag.

Removal of a combination of endocrine disruptors from aqueous systems by seedlings of radish and ryegrass

Endocrine disruptors (EDs) are widespread in the environment, especially aquatic systems, and cause dangerous effects on wildlife and humans. This work was aimed to assess the capacity of radish (Raphanus sativus L.) and ryegrass (Lolium perenne L.) seedlings to tolerate and remove two combinations of EDs containing bisphenol A (BPA), 17α-ethynilestradiol (EE2), and linuron from four aqueous media: distilled water, a solution of natural organic matter (NOM), a lake water and a river water. Seeds of the two species were germinated in each contaminated medium and, at the end of germination, the seedling growth was evaluated by biometric measurements and residual EDs were quantified by chromatographic analysis. Biometric measurements revealed that the phytotoxicity of the two combinations of EDs depended on the medium used. Radish showed a discrete tolerance in distilled water and lake water but was inhibited in the solution of NOM and river water. Ryegrass was negatively affected mainly in river water. The concentration of each ED appeared significantly reduced in all media in the presence of seedlings of both species, but not in the blanks without plants. In 5 days, radish removed up to 88% of BPA, 100% of EE2 and 42% of linuron, and in 6 days ryegrass removed up to 92% of BPA, 74% of EE2 and 16% of linuron. The considerable removal capacity of radish and ryegrass in all media tested encourages the use of phytoremediation to remove EDs from waters. © 2013 Copyright Taylor and Francis Group, LLC.

EFFICIENT FAULT LOCATION IN UNDERGROUND DISTRIBUTION SYSTEMS THROUGH OPTIMUM-PATH FOREST

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Shear bond strength of metal-ceramic repair systems

Statement of problem. When clinical fractures of the ceramic veneer on metal-ceramic prostheses can be repaired, the need for remake may be eliminated or postponed. Many different ceramic repair materials are available, and bond strength data are necessary for predicting the success of a given repair system.Purpose. This study evaluated the shear bond strength of different repair systems for metal-ceramic restorations applied on metal and porcelain.Material and methods. Fifty cylindrical specimens (9 X 3 mm) were fabricated in a nickel-chromium alloy (Vera Bond 11) and 50 in feldspathic porcelain (Noritakc). Metal (M) and porcelain (P) specimens were embedded in a polyvinyl chloride (PVC) ring and received I of the following bonding and resin composite repair systems (n=10): Clearfil SE Bond/Clearfil AP-X (CL), Bistite II DC/Palfique (BT), Cojet Sand/Z100 (Q), Scotchbond Multipurpose Plus/Z100 (SB) (control group), or Cojet Sand plus Scotchbond Multipurpose Plus/Z100 (CJSB). The specimens were stored in distilled water for 24 hours at 37 degrees C, thermal cycled (1000 cycles at 5 degrees C to 55 degrees C), and stored at 37 degrees C for 8 days. Shear bond tests between the metal or ceramic specimens and repair systems were performed in a mechanical testing machine with a crosshead speed of 0.5 mm/min. Mean shear bond strength values (MPa) were submitted to 1-way ANOVA and Tukey honestly significant difference tests (alpha=.05). Each specimen was examined under a stereoscopic lens with X 30 magnification, and mode of failure was classified as adhesive, cohesive, or a combination.Results. on metal, the mean shear bond strength values for the groups were as follows: MCL, 18.40 +/- 2.88(b); MBT, 8.57 +/- 1.00(d); MCJ, 25.24 +/- 3.46(a); MSB, 16.26 +/- 3.09(bc); and MCJSB, 13.11 +/- 1.24(c). on porcelain, the mean shear bond strength values ofeach group were as follows: PCL, 16.91 +/- 2.22(b); PBT, 18.04 +/- 3.2(ab); PCJ, 19.54 +/- 3.77(ab); PSB, 21.05 +/- 3.22(a); and PCJSB, 16.18 +/- 1.71(b). Within each substrate, identical superscript letters denote no significant differences among groups.Conclusions. The bond strength for the metal substrate was significantly higher using the Q system. For porcelain, SB, Q, and BT systems showed the highest shear bond strength values, and only SB was significantly different compared to CL and CJSB (P <.05).

A machine learning approach for genome-wide prediction of morbid and druggable human genes based on systems-level data

Background: The genome-wide identification of both morbid genes, i.e., those genes whose mutations cause hereditary human diseases, and druggable genes, i.e., genes coding for proteins whose modulation by small molecules elicits phenotypic effects, requires experimental approaches that are time-consuming and laborious. Thus, a computational approach which could accurately predict such genes on a genome-wide scale would be invaluable for accelerating the pace of discovery of causal relationships between genes and diseases as well as the determination of druggability of gene products.Results: In this paper we propose a machine learning-based computational approach to predict morbid and druggable genes on a genome-wide scale. For this purpose, we constructed a decision tree-based meta-classifier and trained it on datasets containing, for each morbid and druggable gene, network topological features, tissue expression profile and subcellular localization data as learning attributes. This meta-classifier correctly recovered 65% of known morbid genes with a precision of 66% and correctly recovered 78% of known druggable genes with a precision of 75%. It was than used to assign morbidity and druggability scores to genes not known to be morbid and druggable and we showed a good match between these scores and literature data. Finally, we generated decision trees by training the J48 algorithm on the morbidity and druggability datasets to discover cellular rules for morbidity and druggability and, among the rules, we found that the number of regulating transcription factors and plasma membrane localization are the most important factors to morbidity and druggability, respectively.Conclusions: We were able to demonstrate that network topological features along with tissue expression profile and subcellular localization can reliably predict human morbid and druggable genes on a genome-wide scale. Moreover, by constructing decision trees based on these data, we could discover cellular rules governing morbidity and druggability.

Multiscale Fractal Descriptors and Polynomial Classifier for Partial Pixels Identification in Regions of Interest of Mammographic Images

Computer systems are used to support breast cancer diagnosis, with decisions taken from measurements carried out in regions of interest (ROIs). We show that support decisions obtained from square or rectangular ROIs can to include background regions with different behavior of healthy or diseased tissues. In this study, the background regions were identified as Partial Pixels (PP), obtained with a multilevel method of segmentation based on maximum entropy. The behaviors of healthy, diseased and partial tissues were quantified by fractal dimension and multiscale lacunarity, calculated through signatures of textures. The separability of groups was achieved using a polynomial classifier. The polynomials have powerful approximation properties as classifiers to treat characteristics linearly separable or not. This proposed method allowed quantifying the ROIs investigated and demonstrated that different behaviors are obtained, with distinctions of 90% for images obtained in the Cranio-caudal (CC) and Mediolateral Oblique (MLO) views.

Suppression of vibrations in strongly nonhomogeneous 2DOF systems

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

On energy transfer phenomena, in a nonlinear ideal and nonideal essential vibrating systems, coupled to a (MR) magneto-rheological damper

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Phase Transition in Dynamical Systems: Defining Classes of Universality for Two-Dimensional Hamiltonian Mappings via Critical Exponents

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Efficiency of bioaugmentation in the removal of organic matter in aquaculture systems

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

The influence of internal surface treatments on tensile bond strength for two ceramic systems

Statement of the Problem: the ceramic composition and surface microstructure of all-ceramic restorations are important components of an effective bonding substrate. Hydrofluoric acid and sandblasting are well-known procedures for surface treatment; however, surface treatment for high alumina-containing and lithium disilicate ceramics have not been fully investigated.Purpose: This in vitro study evaluated the tensile bond strength of resin cement to two types of ceramic systems with different surface treatments.Methods and Materials: Thirty specimens of each ceramic system were made according to the manufacturer's instructions and embedded in polyester resin. Specimens of In-Ceram Alumina [1] and IPS Empress 2 [E] were distributed to three groups with differing surface treatments (n=10): sandblasting with 50 jam aluminum oxide (APA); sandblasting with 110 pm aluminum oxide modified with silica particles (ROCATEC System-RS); a combination of sandblasting with APA and 10% hydrofluoric acid etching (HA) for two minutes on In-Ceram and for 20 seconds for IPS Empress 2. After the respective surface treatments, all the specimens were silanated, and Rely-X resin cement was injected onto the ceramic surface and light polymerized. The specimens were stored in distilled water at 37 degrees C for 24 hours and thermally cycled 1,100 times (5 degrees C/55 degrees C). The tensile bond strength test was performed in a universal testing machine at a 0.5 mm/minute crosshead speed.Results: the mean bond strength values (AWa) for IPS Empress 2 were 12.01 +/- 5.93 (EAPA), 10.34 +/- 1.77 (ERS) and 14.49 +/- 3.04 (EHA). The mean bond strength values for In-Ceram Alumina were 9.87 +/- 2.40 JAPA) and 20.40 +/- 6.27 (IRS). All In-Ceram specimens treated with 10% hydrofluoric acid failed during thermal cycling.Conclusion: the Rocatec system was the most effective surface treatment for In-Ceram Alumina ceramics; whereas, the combination of aluminum oxide sandblasting and hydrofluoric acid etching for 20 seconds worked more effectively for Empress 2 ceramics.

ALPHA-CONTRACTIONS AND ATTRACTORS FOR DISSIPATIVE SEMILINEAR HYPERBOLIC-EQUATIONS AND SYSTEMS

In this paper we discuss the existence of compact attractor for the abstract semilinear evolution equation u = Au + f (t, u); the results are applied to damped partial differential equations of hyperbolic type. Our approach is a combination of Liapunov method with the theory of alpha-contractions.

RESTRUCTURING of SYSTEMS and LOGISTICS of MOVEMENT and ITS IMPACT on URBAN and REGIONAL PLANNING PAULISTA

Report some of the changes in production and consumption occurring in the state of São Paulo. through the restructuring in motion systems, logistics and standards and taxation, as well as the impacts on urban spaces through new economic dynamics, imposed by the demands of corporate, is the purpose of this article. The decentralization of production and consumption towards the interior was made possible by the combination of hierarchical and ordered some basic elements such as technological innovations (ways and means of transport) and organizational (logistics, standards and taxation) which optimized the flow territorial state São Paulo. It is noteworthy, therefore: 1) the improvement of logistics as a strategy, planning and management of transport, storage and communications (including the granting of public services to private), 2) the technological improvement and expansion of motion systems (infrastructure, means of transport) and 3) the systems of rules and regulations through taxation and deregulation affect the circulatory system of a given space. Thus, both systems aims to disentangle the economic flows (goods, services, information, capital and people) and provide a more fluid territorial. The impacts on the State of São Paulo, mainly through its economic dynamics, revert positively and negatively, by changing the way one thinks and performs planning.

Methanol laser line assignments: evidence of four-level laser systems

The Ritz computer program, developed for facilitating the assignment of molecular Fourier transform absorption spectra and described in a previous work, determines the energy level values involved in the assigned transitions by the Rydberg-Ritz combination principle. Combining the data obtained from the analyses of high-resolution infrared (IR) and far-infrared (FIR) spectra, it is possible to predict possible FIR laser emissions of molecules. In the present work we have applied this method to the common isotopomer methanol, 12CH3 16OH, and obtained 14 proposed assignments for previously unassigned FIR laser lines. We also predict 15 possible new FIR laser emissions. For the first time, an assignment involving a four-level laser system with collisional population transfer to a slightly higher energy level is reported. © 1998 Academic Press.

Cariostatic effect of fluoride-containing restorative systems associated with dentifrices on root dentin

Objective. The aim of this in vitro study was to evaluate the interaction between two sources of fluoride (restorative systems and dentifrices) in inhibiting artificial root caries development. Methods. One hundred and eighty tooth segments were embedded in polyester resin, and sanded flat. Cylindrical cavities 1.0 mm-deep and 1.5 mm-diameter were prepared in root dentin and randomly restored by fluoride-containing restorative systems: Ketac-fil/Espe (Ke), Fuji II LC/GC Corp (Fj), F2000/3M (F2), Surefil/Dentsply (Su) or a control: Filtek Z250/3M (Z2). Ten experimental groups were made to test the association among the five restorative systems and two dentifrices: with F - (Sensodyne Baking Soda) or without F- (Sensodyne Original) (n = 18). After surface polishing, a 1 mm-wide margin around the restorations was demarcated and initial dentin surface Knoop microhardness values (KHNi) were obtained. The specimens were submitted to a pH-cycling model, and to applications of slurries of dentifrice. Afterwards the final dentin surface Knoop microhardness values (KHNf) were measured. Results. The differences between KHNi and KHNf, and the covariate KHNi were considered by the ANCOVA and Tukey's test (α = 0.05). The interaction between restorative system and dentifrice was statistically significant (p = 0.0026). All restorative systems provided some protection against artificial caries challenge when associated with the fluoride-containing dentifrice treatment. The means (standard deviation) of reductions in Knoop hardness values for systems associated with the fluoride-containing dentifrice were: Ke: 40.0(1.02)a, Fj: 41.9(1.02)b, F2: 43.3(1.04)c, Su: 43.5(1.00)c, Z2: 44.0(1.02)c; and with the non-fluoride-containing dentifrice were: Ke: 42.9(1.02)a, Fj: 44.7(1.01)b, F2: 45.2(1.09)bc, Su: 46.0(0.99)c, Z2: 46.6(0.99)c (statistical differences were expressed by different letters). Conclusion. The cariostatic effect shown by the fluoride-containing dentifrice could enhance that shown by Ketac-fil and Fuji II LC, and could mask that shown by F2000. © 2002 Elsevier Science Ltd. All rights reserved.