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.

Using data mining techniques to support breast cancer diagnosis

More than ever, there is an increase of the number of decision support methods and computer aided diagnostic systems applied to various areas of medicine. In breast cancer research, many works have been done in order to reduce false-positives when used as a double reading method. In this study, we aimed to present a set of data mining techniques that were applied to approach a decision support system in the area of breast cancer diagnosis. This method is geared to assist clinical practice in identifying mammographic findings such as microcalcifications, masses and even normal tissues, in order to avoid misdiagnosis. In this work a reliable database was used, with 410 images from about 115 patients, containing previous reviews performed by radiologists as microcalcifications, masses and also normal tissue findings. Throughout this work, two feature extraction techniques were used: the gray level co-occurrence matrix and the gray level run length matrix. For classification purposes, we considered various scenarios according to different distinct patterns of injuries and several classifiers in order to distinguish the best performance in each case described. The many classifiers used were Naïve Bayes, Support Vector Machines, k-nearest Neighbors and Decision Trees (J48 and Random Forests). The results in distinguishing mammographic findings revealed great percentages of PPV and very good accuracy values. Furthermore, it also presented other related results of classification of breast density and BI-RADS® scale. The best predictive method found for all tested groups was the Random Forest classifier, and the best performance has been achieved through the distinction of microcalcifications. The conclusions based on the several tested scenarios represent a new perspective in breast cancer diagnosis using data mining techniques.

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.

Evolution of IgG antibody response against Toxoplasma gondii tissue cyst in acute and chronic human infections

The recognition profile of the tissue cysts antigens by IgG antibodies was studied during acute and chronic human toxoplasmic infection. Thus the IgG response against Toxoplasma gondii was investigated by immunoblotting in two patients accidentally infected with the RH strain as well as in group of naturally infected patients at acute and chronic phase. There was an overall coincidence of molecular mass among antigens of tachyzoites and tissue cysts recognized by these sera, however, they appear not to be the same molecules. The response against tissue cysts starts early during acute infection, and the reactivity of antibodies is strong against a wide range of antigens. Six bands (between 82 and 151 kDa) were exclusively recognized by chronic phase sera but only the 132 kDa band was positive in more than 50% of the sera analysed. A mixture of these antigens could be used to discriminate between the two infection phases. The most important antigens recognized by the acute and the chronic phase sera were 4 clusters in the ranges 20-24 kDa, 34-39 kDa, 58-80 kDa and 105-130 kDa as well as two additional antigens of 18 and 29 kDa. Both accidentally infected patients and some of the naturally infected patients showed a weak specific response against tissue cyst antigens.

Inoculation of Lacazia loboi into the subcutaneous tissue of the hamster cheek pouch

The subcutaneous tissue of the hamster cheek pouch, a site of immunologic privilege, has been used to investigate the potential infectivity of different types of parasites. It has been demonstrated that the implantation of fragments of lesions induced by the fungus Lacazia loboi, the etiologic agent of Jorge Lobo's disease, into the subcutaneous tissue of the hamster cheek pouch resulted in parasite multiplication and dissemination to satellite lymph nodes16. Here we describe the evolution of lesions induced by the inoculation of the isolated fungus into this immunologically privileged site. The morphology of the inflammatory response and fungal viability and proliferation were evaluated. Inoculation of the fungus into the cheek pouch induced histiocytic granulomas with rare lymphocytes. Although fungal cells were detected for a period of up to 180 days in these lesions, the fungi lost viability after the first day of inoculation. In contrast, when the parasite was inoculated into the footpad, non-organized histiocytic lesions were observed. Langhan's giant cells, lymphocytes and fungal particles were observed in these lesions. Fungal viability was observed up to 60 days after inoculation and non-viable parasites were present in the persistent lesions up to 180 days post-inoculation. These data indicate that the subcutaneous tissue of the hamster cheek pouch is not a suitable site for the proliferation of Lacazia loboi when the fungus isolated from human tissues is tested.

Immune response and tissue cytoprotection: two sides of the same coin in immunopathology

Dissertation presented to obtain the Ph.D degree in Biochemistry

The histopathology of cutaneous leishmaniasis due to Leishmania (Leishmania) mexicana in the Yucatan peninsula, Mexico

Localized Cutaneous Leishmaniasis (LCL) known as "chiclero's ulcer" in southeast Mexico, was described by SEIDELIN in 1912. Since then the sylvatic region of the Yucatan peninsula has been documented as an endemic focus of LCL. This study of 73 biopsies from parasitological confirmed lesions of LCL cases of Leishmania (Leishmania) mexicana infection was undertaken: 1) to examine host response at tissue level; and 2) to relate manifestations of this response to some characteristics of clinical presentation. Based on Magalhães' classification we found that the most common pattern in our LCL cases caused by L. (L.) mexicana was predominantly characterized by the presence of unorganized granuloma without necrosis, (43.8%). Another important finding to be highlighted is the fact that in 50/73 (68.5%) parasite identification was positive. There was direct relation between the size of the lesion and time of evolution (r s = 0.3079, p = 0.03), and inverse correlation between size of the lesion and abundance of amastigotes (r s = -0.2467, p = 0.03). In view of the complexity of clinical and histopathological findings, cell-mediated immune response of the disease related to clinical and histopathological features, as so genetic background should be studied.

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