Network-based high level data classification

Traditional supervised data classification considers only physical features (e. g., distance or similarity) of the input data. Here, this type of learning is called low level classification. On the other hand, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is, here, referred to as high level classification. In this paper, we propose a hybrid classification technique that combines both types of learning. The low level term can be implemented by any classification technique, while the high level term is realized by the extraction of features of the underlying network constructed from the input data. Thus, the former classifies the test instances by their physical features or class topologies, while the latter measures the compliance of the test instances to the pattern formation of the data. Our study shows that the proposed technique not only can realize classification according to the pattern formation, but also is able to improve the performance of traditional classification techniques. Furthermore, as the class configuration's complexity increases, such as the mixture among different classes, a larger portion of the high level term is required to get correct classification. This feature confirms that the high level classification has a special importance in complex situations of classification. Finally, we show how the proposed technique can be employed in a real-world application, where it is capable of identifying variations and distortions of handwritten digit images. As a result, it supplies an improvement in the overall pattern recognition rate.

Semi-supervised dimensionality reduction based on partial least squares for visual analysis of high dimensional data

Dimensionality reduction is employed for visual data analysis as a way to obtaining reduced spaces for high dimensional data or to mapping data directly into 2D or 3D spaces. Although techniques have evolved to improve data segregation on reduced or visual spaces, they have limited capabilities for adjusting the results according to user's knowledge. In this paper, we propose a novel approach to handling both dimensionality reduction and visualization of high dimensional data, taking into account user's input. It employs Partial Least Squares (PLS), a statistical tool to perform retrieval of latent spaces focusing on the discriminability of the data. The method employs a training set for building a highly precise model that can then be applied to a much larger data set very effectively. The reduced data set can be exhibited using various existing visualization techniques. The training data is important to code user's knowledge into the loop. However, this work also devises a strategy for calculating PLS reduced spaces when no training data is available. The approach produces increasingly precise visual mappings as the user feeds back his or her knowledge and is capable of working with small and unbalanced training sets.

Isotopic, trace element and nutrient characterization of coastal waters from Ubatuba inner shelf area, south-eastern Brazil

Stable isotopes, tritium, radium isotopes, radon, trace elements and nutrients data were collected during two sampling campaigns in the Ubatuba coastal area (south-eastern Brazil) with the aim of investigating submarine groundwater discharge (SGD) in the region. The isotopic composition (delta D, delta(18)O, (3)H) of submarine waters was characterised by significant variability and heavy isotope enrichment. The stable isotopes and tritium data showed good separation of groundwater and seawater groups. The contribution of groundwater in submarine waters varied from a few % to 17%. Spatial distribution of (222)Rn activity concentration in surface seawater revealed changes between 50 and 200 Bq m(-3) which were in opposite relationship with observed salinities. Time series measurements of (222)Rn activity concentration in Flamengo Bay (from 1 to 5 kBq m(-3)), obtained by in situ underwater gamma-spectrometry showed a negative correlation between the (222)Rn activity concentration and tide/salinity. This may be caused by sea level changes as tide effects induce variations of hydraulic gradients, which increase (222)Rn concentration during lower sea level, and opposite, during high tides where the (222)Rn activity concentration is smaller. The estimated SGD fluxes varied during 22-26 November between 8 and 40 cm d(-1), with an average value of 21 cm d(-1) (the unit is cm(3)/cm(2) per day). The radium isotopes and nutrient data showed scattered distributions with offshore distance and salinity. which implies that in a complex coast with many small bays and islands, the area has been influenced by local currents and groundwater-seawater mixing. SGD in the Ubatuba area is fed by coastal contaminated groundwater and re-circulated seawater (with small admixtures of groundwater). which claims for potential environmental concern with implications on the management of freshwater resources in the region. (C) 2007 Elsevier Ltd. All rights reserved.

LTR point mutations in the Tax-responsive elements of HTLV-1 isolates from HIV/HTLV-1-coinfected patients

Background: In Virology Journal 2011, 8: 535, Neto et al. described point mutations into Tax-responsive elements (TRE) of the LTR region of HTLV-1 isolates from asymptomatic carriers from Sao Paulo, Brazil, and hypothesized that the presence of the G232A mutation in the TRE-1 increase viral proliferation and consequently the proviral load (PvL), while the A184G mutation in the TRE-2 do not have such effect. Findings: We performed the real-time PCR assay (pol) and sequenced LTR region of HTLV-1 isolates from 24 HIV/HTLV-1-coinfected patients without HTLV-1-associated diseases from the same geographic area. These sequences were classified as belonging to the transcontinental subgroup A of the Cosmopolitan subtype a. The frequency of G232A mutation (16/24, 66.7%) was high as much as 61.8% reported by Neto's in HTLV-1 asymptomatic carriers with high PvL. High frequency (13/24, 54.2%) of double mutations G232A and A184G was also detected in HIV/HTLV-1-coinfected patients. We did not quantify PvL, but comparative analyses of the cycle threshold (Ct) median values of the group of isolates presenting the mutated-types sequences (Ct 33.5, n = 16) versus the group of isolates with the wild-type sequences (Ct 32, n = 8) showed no statistical difference (p = 0.4220). Conclusion: The frequencies of mutated-type sequences in the TRE-1 and TRE-2 motifs were high in HIV/HTLV-1-coinfected patients from Sao Paulo, Brazil. If these LTR point mutations have predictive value for the development of HTLV-1-associated diseases or they correspond to the subtype of virus that circulate in this geographic area has to be determined.