Classification Performance of Multilayer Perceptrons with Different Risk Functionals

In the present paper we assess the performance of information-theoretic inspired risks functionals in multilayer perceptrons with reference to the two most popular ones, Mean Square Error and Cross-Entropy. The information-theoretic inspired risks, recently proposed, are: HS and HR2 are, respectively, the Shannon and quadratic Rényi entropies of the error; ZED is a risk reflecting the error density at zero errors; EXP is a generalized exponential risk, able to mimic a wide variety of risk functionals, including the information-thoeretic ones. The experiments were carried out with multilayer perceptrons on 35 public real-world datasets. All experiments were performed according to the same protocol. The statistical tests applied to the experimental results showed that the ubiquitous mean square error was the less interesting risk functional to be used by multilayer perceptrons. Namely, mean square error never achieved a significantly better classification performance than competing risks. Cross-entropy and EXP were the risks found by several tests to be significantly better than their competitors. Counts of significantly better and worse risks have also shown the usefulness of HS and HR2 for some datasets.

Automatic cymbal classification

Dissertação apresentada na Faculdade de Ciências e Tecnologia da Universidade Nova de Lisboa para a obtenção do grau de Mestre em Engenharia Informática

Partial classification of Lorenz knots: Syllable permutations of torus knots words

We define families of aperiodic words associated to Lorenz knots that arise naturally as syllable permutations of symbolic words corresponding to torus knots. An algorithm to construct symbolic words of satellite Lorenz knots is defined. We prove, subject to the validity of a previous conjecture, that Lorenz knots coded by some of these families of words are hyperbolic, by showing that they are neither satellites nor torus knots and making use of Thurston's theorem. Infinite families of hyperbolic Lorenz knots are generated in this way, to our knowledge, for the first time. The techniques used can be generalized to study other families of Lorenz knots.

Text classification using compression-based dissimilarity measures

Arguably, the most difficult task in text classification is to choose an appropriate set of features that allows machine learning algorithms to provide accurate classification. Most state-of-the-art techniques for this task involve careful feature engineering and a pre-processing stage, which may be too expensive in the emerging context of massive collections of electronic texts. In this paper, we propose efficient methods for text classification based on information-theoretic dissimilarity measures, which are used to define dissimilarity-based representations. These methods dispense with any feature design or engineering, by mapping texts into a feature space using universal dissimilarity measures; in this space, classical classifiers (e.g. nearest neighbor or support vector machines) can then be used. The reported experimental evaluation of the proposed methods, on sentiment polarity analysis and authorship attribution problems, reveals that it approximates, sometimes even outperforms previous state-of-the-art techniques, despite being much simpler, in the sense that they do not require any text pre-processing or feature engineering.

Multi-scale vegetation classification using earth observation data of the Sundarban mangrove forest, Bangladesh.

Dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Picocyanobacteria From a Clade Of Marine Cyanobium Revealed Bioactive Potential Against Microalgae, Bacteria, And Marine

The production of bioactive compounds either toxic or with pharmacological applications by cyanobacteria is well established. However, picoplanktonic forms within this group of organisms have rarely been studied in this context. In this study, the toxicological potential of picocyanobacteria from a clade of marine Cyanobium strains isolated from the Portuguese coast was examined using different biological models. First, strains were identified by applying morphological and molecular approaches and cultured under lab conditions. A crude extract and three fractions reflecting a preliminary segregation of lipophilic metabolites were tested for toxicity with the marine microalga Nannochloropsis sp., the bacteria Pseudomonas sp., the brine shrimp Artemia salina, and fertilized eggs of the sea urchin Paracentrotus lividus. No significant apparent adverse effects were noted against Artemia salina. However, significant adverse effects were found in all other assays, with an inhibition of Nannochloropsis sp. and Pseudomonas sp. growth and marked reduction in Paracentrotus lividus larvae length. The results obtained indicated that Cyanobium genus may serve as a potential source of interesting bioactive compounds and emphasize the importance of also studying smaller picoplanktonic fractions of marine cyanobacteria.

Toxicity assessment of crude and partially purified extracts of marine Synechocystis and Synechococcus cyanobacterial strains in marine invertebrates

Among the Cyanoprokaryota, the genera Synechocystis and Synechococcus have rarely been studied with respect to potential toxicity. This is particularly true with marine environments where studies about the toxicity of cyanobacteria are restricted to filamentous forms at the warmer temperate and tropical regions and also to filamentous forms at cold seas such as the Baltic Sea. In this study, we describe the effects of cyanobacterial strains of the Synechocystis and Synechococcus genera isolated from the marine coast of Portugal, on marine invertebrates. Crude and partially purified extracts at a concentration of 100 mg/ml of freeze-dried material of the marine strains were tested for acute toxicity in nauplii of the brine shrimp Artemia salina, in the rotifer Brachionus plicatillis and in embryos of the sea urchin Paracentrotus lividus and the mussel Mytilus galloprovincialis. The cyanobacterial extracts, especially the crude extract, had an impact on A. salina nauplii. No significant toxic effects were registered against the rotifer. A negative impact of all strains was recorded on the embryonic development of the sea urchin, with toxic effects resulting in an inhibition of embryogenesis or development of smaller larvae. To the mussel embryos, the effects of cyanobacterial extracts resulted in a complete inhibition of embryogenesis. The results of all assays indicate that Synechocystis and Synechococcus marine strains contained toxic compounds to marine invertebrates.

Marine Operations with the SWORDFISH Autonomous Surface Vehicle

IEEE Robótica 2007 - 7th Conference on Mobile Robots and Competitions, Paderne, Portugal 2007

On the Sensor Classification Scheme of Robotic Manipulators

This paper analyzes the signals captured during impacts and vibrations of a mechanical manipulator. The Fourier Transform of eighteen different signals are calculated and approximated by trendlines based on a power law formula. A sensor classification scheme based on the frequency spectrum behavior is presented.

Towards a classification scheme for musical sounds

Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA), 2013

Field experiments for marine casualty detection with autonomous surface vehicles

In this paper we present a set of field tests for detection of human in the water with an unmanned surface vehicle using infrared and color cameras. These experiments aimed to contribute in the development of victim target tracking and obstacle avoidance for unmanned surface vehicles operating in marine search and rescue missions. This research is integrated in the work conducted in the European FP7 research project Icarus aiming to develop robotic tools for large scale rescue operations. The tests consisted in the use of the ROAZ unmanned surface vehicle equipped with a precision GPS system for localization and both visible spectrum and IR cameras to detect the target. In the experimental setup, the test human target was deployed in the water wearing a life vest and a diver suit (thus having lower temperature signature in the body except hands and head) and was equipped with a GPS logger. Multiple target approaches were performed in order to test the system with different sun incidence relative angles. The experimental setup, detection method and preliminary results from the field trials performed in the summer of 2013 in Sesimbra, Portugal and in La Spezia, Italy are also presented in this work.

Thermographic and visible spectrum camera calibration for marine robotic target detection

Oceans - San Diego, 2013

Assessment of non-steroidal anti-inflammatory and analgesic pharmaceuticals in seawaters of North of Portugal: Occurrence and environmental risk

The occurrence of seven pharmaceuticals and two metabolites belonging to non-steroidal anti-inflammatory drugs and analgesics therapeutic classes was studied in seawaters. A total of 101 samples covering fourteen beaches and five cities were evaluated in order to assess the spatial distribution of pharmaceuticals among north Portuguese coast. Seawaters were selected in order to embrace different bathing water quality (excellent, good and sufficient). Acetaminophen, ketoprofen and the metabolite hydroxyibuprofen were detected in all the seawater samples at maximum concentrations of 584, 89.7 and 287 ng L− 1, respectively. Carboxyibuprofen had the highest seawater concentration (1227 ng L− 1). The temporal distribution of the selected pharmaceuticals during the bathing season showed that, in general, higher concentrations were detected in August and September. The environmental risk posed by the pharmaceuticals detected in seawaters towards different trophic levels (fish, daphnids and algae) was also assessed. Only diclofenac showed hazard quotients above one for fish, representing a potential risk for aquatic organisms. These results were observed in seawaters classified as excellent bathing water. Additional data is needed in order to support the identification and prioritization of risks posed by pharmaceuticals in marine environment.

Marine Cyanobacteria Compounds with Anticancer Properties: Implication of Apoptosis

Marine cyanobacteria have been proved to be an important source of potential anticancer drugs. Although several compounds were found to be cytotoxic to cancer cells in culture, the pathways by which cells are affected are still poorly elucidated. For some compounds, cancer cell death was attributed to an implication of apoptosis through morphological apoptotic features, implication of caspases and proteins of the Bcl-2 family, and other mechanisms such as interference with microtubules dynamics, cell cycle arrest and inhibition of proteases other than caspases.

Impact of RIFLE Classification in Liver Transplantation

Acute renal failure (ARF) is common after orthotopic liver transplantation (OLT). The aim of this study was to evaluate the prognostic value of RIFLE classification in the development of CKD, hemodialysis requirement, and mortality. Patients were categorized as risk (R), injury (I) or failure (F) according to renal function at day 1, 7 and 21. Final renal function was classified according to K/DIGO guidelines. We studied 708 OLT recipients, transplanted between September 1992 and March 2007; mean age 44 +/- 12.6 yr, mean follow-up 3.6 yr (28.8% > or = 5 yr). Renal dysfunction before OLT was known in 21.6%. According to the RIFLE classification, ARF occurred in 33.2%: 16.8% were R class, 8.5% I class and 7.9% F class. CKD developed in 45.6%, with stages 4 or 5d in 11.3%. Mortality for R, I and F classes were, respectively, 10.9%, 13.3% and 39.3%. Severity of ARF correlated with development of CKD: stage 3 was associated with all classes of ARF, stages 4 and 5d only with severe ARF. Hemodialysis requirement (23%) and mortality were only correlated with the most severe form of ARF (F class). In conclusion, RIFLE classification is a useful tool to stratify the severity of early ARF providing a prognostic indicator for the risk of CKD occurrence and death.