A methodology to compare dimensionality reduction algorithms in terms of loss of quality

Dimensionality Reduction (DR) is attracting more attention these days as a result of the increasing need to handle huge amounts of data effectively. DR methods allow the number of initial features to be reduced considerably until a set of them is found that allows the original properties of the data to be kept. However, their use entails an inherent loss of quality that is likely to affect the understanding of the data, in terms of data analysis. This loss of quality could be determinant when selecting a DR method, because of the nature of each method. In this paper, we propose a methodology that allows different DR methods to be analyzed and compared as regards the loss of quality produced by them. This methodology makes use of the concept of preservation of geometry (quality assessment criteria) to assess the loss of quality. Experiments have been carried out by using the most well-known DR algorithms and quality assessment criteria, based on the literature. These experiments have been applied on 12 real-world datasets. Results obtained so far show that it is possible to establish a method to select the most appropriate DR method, in terms of minimum loss of quality. Experiments have also highlighted some interesting relationships between the quality assessment criteria. Finally, the methodology allows the appropriate choice of dimensionality for reducing data to be established, whilst giving rise to a minimum loss of quality.

Reducing the dimensionality effect in importance sampling simulations

The dimensionality effect is avoided by the use of sufficient statistics in event probability estimators realised by importance sampling. If the system function is not a sufficient statistic, an approach is proposed to reduce the dimensionality effect in the estimators. Simulation results of false-alarm probability estimations, applied to radar detection, confirm a clear concordance with the theoretical results

A visual framework to accelerate knowledge discovery based on dimensionality reduction minimizing degradation of quality

Tradicionalmente, el uso de técnicas de análisis de datos ha sido una de las principales vías para el descubrimiento de conocimiento oculto en grandes cantidades de datos, recopilados por expertos en diferentes dominios. Por otra parte, las técnicas de visualización también se han usado para mejorar y facilitar este proceso. Sin embargo, existen limitaciones serias en la obtención de conocimiento, ya que suele ser un proceso lento, tedioso y en muchas ocasiones infructífero, debido a la dificultad de las personas para comprender conjuntos de datos de grandes dimensiones. Otro gran inconveniente, pocas veces tenido en cuenta por los expertos que analizan grandes conjuntos de datos, es la degradación involuntaria a la que someten a los datos durante las tareas de análisis, previas a la obtención final de conclusiones. Por degradación quiere decirse que los datos pueden perder sus propiedades originales, y suele producirse por una reducción inapropiada de los datos, alterando así su naturaleza original y llevando en muchos casos a interpretaciones y conclusiones erróneas que podrían tener serias implicaciones. Además, este hecho adquiere una importancia trascendental cuando los datos pertenecen al dominio médico o biológico, y la vida de diferentes personas depende de esta toma final de decisiones, en algunas ocasiones llevada a cabo de forma inapropiada. Ésta es la motivación de la presente tesis, la cual propone un nuevo framework visual, llamado MedVir, que combina la potencia de técnicas avanzadas de visualización y minería de datos para tratar de dar solución a estos grandes inconvenientes existentes en el proceso de descubrimiento de información válida. El objetivo principal es hacer más fácil, comprensible, intuitivo y rápido el proceso de adquisición de conocimiento al que se enfrentan los expertos cuando trabajan con grandes conjuntos de datos en diferentes dominios. Para ello, en primer lugar, se lleva a cabo una fuerte disminución en el tamaño de los datos con el objetivo de facilitar al experto su manejo, y a la vez preservando intactas, en la medida de lo posible, sus propiedades originales. Después, se hace uso de efectivas técnicas de visualización para representar los datos obtenidos, permitiendo al experto interactuar de forma sencilla e intuitiva con los datos, llevar a cabo diferentes tareas de análisis de datos y así estimular visualmente su capacidad de comprensión. De este modo, el objetivo subyacente se basa en abstraer al experto, en la medida de lo posible, de la complejidad de sus datos originales para presentarle una versión más comprensible, que facilite y acelere la tarea final de descubrimiento de conocimiento. MedVir se ha aplicado satisfactoriamente, entre otros, al campo de la magnetoencefalografía (MEG), que consiste en la predicción en la rehabilitación de lesiones cerebrales traumáticas (Traumatic Brain Injury (TBI) rehabilitation prediction). Los resultados obtenidos demuestran la efectividad del framework a la hora de acelerar y facilitar el proceso de descubrimiento de conocimiento sobre conjuntos de datos reales. ABSTRACT Traditionally, the use of data analysis techniques has been one of the main ways of discovering knowledge hidden in large amounts of data, collected by experts in different domains. Moreover, visualization techniques have also been used to enhance and facilitate this process. However, there are serious limitations in the process of knowledge acquisition, as it is often a slow, tedious and many times fruitless process, due to the difficulty for human beings to understand large datasets. Another major drawback, rarely considered by experts that analyze large datasets, is the involuntary degradation to which they subject the data during analysis tasks, prior to obtaining the final conclusions. Degradation means that data can lose part of their original properties, and it is usually caused by improper data reduction, thereby altering their original nature and often leading to erroneous interpretations and conclusions that could have serious implications. Furthermore, this fact gains a trascendental importance when the data belong to medical or biological domain, and the lives of people depends on the final decision-making, which is sometimes conducted improperly. This is the motivation of this thesis, which proposes a new visual framework, called MedVir, which combines the power of advanced visualization techniques and data mining to try to solve these major problems existing in the process of discovery of valid information. Thus, the main objective is to facilitate and to make more understandable, intuitive and fast the process of knowledge acquisition that experts face when working with large datasets in different domains. To achieve this, first, a strong reduction in the size of the data is carried out in order to make the management of the data easier to the expert, while preserving intact, as far as possible, the original properties of the data. Then, effective visualization techniques are used to represent the obtained data, allowing the expert to interact easily and intuitively with the data, to carry out different data analysis tasks, and so visually stimulating their comprehension capacity. Therefore, the underlying objective is based on abstracting the expert, as far as possible, from the complexity of the original data to present him a more understandable version, thus facilitating and accelerating the task of knowledge discovery. MedVir has been succesfully applied to, among others, the field of magnetoencephalography (MEG), which consists in predicting the rehabilitation of Traumatic Brain Injury (TBI). The results obtained successfully demonstrate the effectiveness of the framework to accelerate and facilitate the process of knowledge discovery on real world datasets.

Analysis of pattern recognition and dimensionality reduction techniques for odor biometrics

In this paper, we analyze the performance of several well-known pattern recognition and dimensionality reduction techniques when applied to mass-spectrometry data for odor biometric identification. Motivated by the successful results of previous works capturing the odor from other parts of the body, this work attempts to evaluate the feasibility of identifying people by the odor emanated from the hands. By formulating this task according to a machine learning scheme, the problem is identified with a small-sample-size supervised classification problem in which the input data is formed by mass spectrograms from the hand odor of 13 subjects captured in different sessions. The high dimensionality of the data makes it necessary to apply feature selection and extraction techniques together with a simple classifier in order to improve the generalization capabilities of the model. Our experimental results achieve recognition rates over 85% which reveals that there exists discriminatory information in the hand odor and points at body odor as a promising biometric identifier.

Dimensionality Reduction Through Sub-Space Mapping for Nearest Neighbour Algorithms

Postprint

(Tables 1-3) Water chemistry of cloud forest streams at baseflow conditions, Rio San Francisco, Ecuador

We investigated controls on the water chemistry of a South Ecuadorian cloud forest catchment which is partly pristine, and partly converted to extensive pasture. From April 2007 to May 2008 water samples were taken weekly to biweekly at nine different subcatchments, and were screened for differences in electric conductivity, pH, anion, as well as element composition. A principal component analysis was conducted to reduce dimensionality of the data set and define major factors explaining variation in the data. Three main factors were isolated by a subset of 10 elements (Ca2+, Ce, Gd, K+, Mg2+, Na+, Nd, Rb, Sr, Y), explaining around 90% of the data variation. Land-use was the major factor controlling and changing water chemistry of the subcatchments. A second factor was associated with the concentration of rare earth elements in water, presumably highlighting other anthropogenic influences such as gravel excavation or road construction. Around 12% of the variation was explained by the third component, which was defined by the occurrence of Rb and K and represents the influence of vegetation dynamics on element accumulation and wash-out. Comparison of base- and fast flow concentrations led to the assumption that a significant portion of soil water from around 30 cm depth contributes to storm flow, as revealed by increased rare earth element concentrations in fast flow samples. Our findings demonstrate the utility of multi-tracer principal component analysis to study tropical headwater streams, and emphasize the need for effective land management in cloud forest catchments.

Aspects of Motives: Finite-dimensionality, Chow-Kunneth Decompositions and Intersections of Cycles

This thesis analyzes the Chow motives of 3 types of smooth projective varieties: the desingularized elliptic self fiber product, the Fano surface of lines on a cubic threefold and an ample hypersurface of an Abelian variety. For the desingularized elliptic self fiber product, we use an isotypic decomposition of the motive to deduce the Murre conjectures. We also prove a result about the intersection product. For the Fano surface of lines, we prove the finite-dimensionality of the Chow motive. Finally, we prove that an ample hypersurface on an Abelian variety possesses a Chow-Kunneth decomposition for which a motivic version of the Lefschetz hyperplane theorem holds.

Effect Of Platform Connection And Abutment Material On Stress Distribution In Single Anterior Implant-supported Restorations: A Nonlinear 3-dimensional Finite Element Analysis.

Although various abutment connections and materials have recently been introduced, insufficient data exist regarding the effect of stress distribution on their mechanical performance. The purpose of this study was to investigate the effect of different abutment materials and platform connections on stress distribution in single anterior implant-supported restorations with the finite element method. Nine experimental groups were modeled from the combination of 3 platform connections (external hexagon, internal hexagon, and Morse tapered) and 3 abutment materials (titanium, zirconia, and hybrid) as follows: external hexagon-titanium, external hexagon-zirconia, external hexagon-hybrid, internal hexagon-titanium, internal hexagon-zirconia, internal hexagon-hybrid, Morse tapered-titanium, Morse tapered-zirconia, and Morse tapered-hybrid. Finite element models consisted of a 4×13-mm implant, anatomic abutment, and lithium disilicate central incisor crown cemented over the abutment. The 49 N occlusal loading was applied in 6 steps to simulate the incisal guidance. Equivalent von Mises stress (σvM) was used for both the qualitative and quantitative evaluation of the implant and abutment in all the groups and the maximum (σmax) and minimum (σmin) principal stresses for the numerical comparison of the zirconia parts. The highest abutment σvM occurred in the Morse-tapered groups and the lowest in the external hexagon-hybrid, internal hexagon-titanium, and internal hexagon-hybrid groups. The σmax and σmin values were lower in the hybrid groups than in the zirconia groups. The stress distribution concentrated in the abutment-implant interface in all the groups, regardless of the platform connection or abutment material. The platform connection influenced the stress on abutments more than the abutment material. The stress values for implants were similar among different platform connections, but greater stress concentrations were observed in internal connections.

Genetic Predictors Of Long-term Response To Growth Hormone (gh) Therapy In Children With Gh Deficiency And Turner Syndrome: The Influence Of A Socs2 Polymorphism.

There is great interindividual variability in the response to GH therapy. Ascertaining genetic factors can improve the accuracy of growth response predictions. Suppressor of cytokine signaling (SOCS)-2 is an intracellular negative regulator of GH receptor (GHR) signaling. The objective of the study was to assess the influence of a SOCS2 polymorphism (rs3782415) and its interactive effect with GHR exon 3 and -202 A/C IGFBP3 (rs2854744) polymorphisms on adult height of patients treated with recombinant human GH (rhGH). Genotypes were correlated with adult height data of 65 Turner syndrome (TS) and 47 GH deficiency (GHD) patients treated with rhGH, by multiple linear regressions. Generalized multifactor dimensionality reduction was used to evaluate gene-gene interactions. Baseline clinical data were indistinguishable among patients with different genotypes. Adult height SD scores of patients with at least one SOCS2 single-nucleotide polymorphism rs3782415-C were 0.7 higher than those homozygous for the T allele (P < .001). SOCS2 (P = .003), GHR-exon 3 (P= .016) and -202 A/C IGFBP3 (P = .013) polymorphisms, together with clinical factors accounted for 58% of the variability in adult height and 82% of the total height SD score gain. Patients harboring any two negative genotypes in these three different loci (homozygosity for SOCS2 T allele; the GHR exon 3 full-length allele and/or the -202C-IGFBP3 allele) were more likely to achieve an adult height at the lower quartile (odds ratio of 13.3; 95% confidence interval of 3.2-54.2, P = .0001). The SOCS2 polymorphism (rs3782415) has an influence on the adult height of children with TS and GHD after long-term rhGH therapy. Polymorphisms located in GHR, IGFBP3, and SOCS2 loci have an influence on the growth outcomes of TS and GHD patients treated with rhGH. The use of these genetic markers could identify among rhGH-treated patients those who are genetically predisposed to have less favorable outcomes.

Synthesis And Antiproliferative Activity Of 8-hydroxyquinoline Derivatives Containing A 1,2,3-triazole Moiety.

Twelve novel 8-hydroxyquinoline derivatives were synthesized with good yields by performing copper-catalyzed Huisgen 1,3-dipolar cycloaddition (click reaction) between an 8-O-alkylated-quinoline containing a terminal alkyne and various aromatic or protected sugar azides. These compounds were evaluated in vitro for their antiproliferative activity on various cancer cell types. Protected sugar derivative 16 was the most active compound in the series, exhibiting potent antiproliferative activity and high selectivity toward ovarian cancer cells (OVCAR-03, GI50 < 0.25 μg mL(-1)); this derivative was more active than the reference drug doxorubicin (OVCAR-03, GI50 = 0.43 μg mL(-1)). In structure-activity relationship (SAR) studies, the physico-chemical parameters of the compounds were evaluated and docking calculations were performed for the α-glucosidase active site to predict the possible mechanism of action of this series of compounds.

Stereoselective Synthesis Of Aryl Cyclopentene Scaffolds By Heck-matsuda Desymmetrization Of 3-cyclopentenol.

A new enantioselective Heck-Matsuda desymmetrization reaction was accomplished by using 3-cyclopentenol to produce chiral five-membered 4-aryl cyclopentenol scaffolds in good yields and high ee's, together with some 3-aryl-cyclopentanones as minor products. Mechanistically, the hydroxyl group of 3-cyclopentenol acts as a directing group and is responsible for the cis- arrangement in the formation of the 4-aryl-cyclopentenols.

Indole-3-carbinol Attenuates The Deleterious Gestational Effects Of Bisphenol A Exposure On The Prostate Gland Of Male F1 Rats.

Bisphenol A (BPA) is a chemical that has been investigated for it potential to cause prostate diseases. In this study, pregnant Sprague-Dawley rats were treated with 25 or 250 μg/kg BPA from gestational day (GD) 10 to GD21 with or without concurrent indole-3-carbinol (I3C) feeding. I3C is a phytochemical, and it affords chemoprotection against many types of neoplasia. Male F1 rats from different litters were euthanized on post-natal day (PND) 21 and PND180. BPA-treated groups showed a significant increase in histopathological lesions, but I3C feeding reversed many of these changes, mainly at PND180. Maternal I3C feeding increased prostate epithelial apoptosis in the BPA-treated groups and across age groups. Furthermore, I3C induced partial normalization of the prostate histoarchitecture. The results pointed to a protective effect of maternal I3C feeding during pregnancy in the BPA-exposed male offspring, thereby indicating reduction in the harmful effects of gestational BPA imprinting on the prostate.

Crystal Structure Of (3e)-3-[(4-nitro-phen-oxy)-meth-yl]-4-phenyl-but-3-en-2-one.

In the title compound, C17H15NO4, the conformation about the C=C double bond [1.348 (2) Å] is E with the ketone group almost co-planar [C-C-C-C torsion angle = 7.2 (2)°] but the phenyl group twisted away [C-C-C-C = 160.93 (17)°]. The terminal aromatic rings are almost perpendicular to each other [dihedral angle = 81.61 (9)°] giving the mol-ecule an overall U-shape. The crystal packing feature benzene-C-H⋯O(ketone) contacts that lead to supra-molecular helical chains along the b axis. These are connected by π-π inter-actions between benzene and phenyl rings [inter-centroid distance = 3.6648 (14) Å], resulting in the formation of a supra-molecular layer in the bc plane.

Crystal Structure Of (3e)-3-(2,4-di-nitro-phen-oxy-meth-yl)-4-phenyl-but-3-en-2-one.

In the title compound, C17H14N2O6, the conformation about the C=C double bond [1.345 (2) Å] is E, with the ketone moiety almost coplanar [C-C-C-C torsion angle = 9.5 (2)°] along with the phenyl ring [C-C-C-C = 5.9 (2)°]. The aromatic rings are almost perpendicular to each other [dihedral angle = 86.66 (7)°]. The 4-nitro moiety is approximately coplanar with the benzene ring to which it is attached [O-N-C-C = 4.2 (2)°], whereas the one in the ortho position is twisted [O-N-C-C = 138.28 (13)°]. The mol-ecules associate via C-H⋯O inter-actions, involving both O atoms from the 2-nitro group, to form a helical supra-molecular chain along [010]. Nitro-nitro N⋯O inter-actions [2.8461 (19) Å] connect the chains into layers that stack along [001].

Galectin-3 Up-regulation In Hypoxic And Nutrient Deprived Microenvironments Promotes Cell Survival.

Galectin-3 (gal-3) is a β-galactoside binding protein related to many tumoral aspects, e.g. angiogenesis, cell growth and motility and resistance to cell death. Evidence has shown its upregulation upon hypoxia, a common feature in solid tumors such as glioblastoma multiformes (GBM). This tumor presents a unique feature described as pseudopalisading cells, which accumulate large amounts of gal-3. Tumor cells far from hypoxic/nutrient deprived areas express little, if any gal-3. Here, we have shown that the hybrid glioma cell line, NG97ht, recapitulates GBM growth forming gal-3 positive pseudopalisades even when cells are grafted subcutaneously in nude mice. In vitro experiments were performed exposing these cells to conditions mimicking tumor areas that display oxygen and nutrient deprivation. Results indicated that gal-3 transcription under hypoxic conditions requires previous protein synthesis and is triggered in a HIF-1α and NF-κB dependent manner. In addition, a significant proportion of cells die only when exposed simultaneously to hypoxia and nutrient deprivation and demonstrate ROS induction. Inhibition of gal-3 expression using siRNA led to protein knockdown followed by a 1.7-2.2 fold increase in cell death. Similar results were also found in a human GBM cell line, T98G. In vivo, U87MG gal-3 knockdown cells inoculated subcutaneously in nude mice demonstrated decreased tumor growth and increased time for tumor engraftment. These results indicate that gal-3 protected cells from cell death under hypoxia and nutrient deprivation in vitro and that gal-3 is a key factor in tumor growth and engraftment in hypoxic and nutrient-deprived microenvironments. Overexpression of gal-3, thus, is part of an adaptive program leading to tumor cell survival under these stressing conditions.