Fitness assessment of document model

Document classification is a supervised machine learning process, where predefined category labels are assigned to documents based on the hypothesis derived from training set of labelled documents. Documents cannot be directly interpreted by a computer system unless they have been modelled as a collection of computable features. Rogati and Yang [M. Rogati and Y. Yang, Resource selection for domain-specific cross-lingual IR, in SIGIR 2004: Proceedings of the 27th annual international conference on Research and Development in Information Retrieval, ACM Press, Sheffied: United Kingdom, pp. 154-161.] pointed out that the effectiveness of document classification system may vary in different domains. This implies that the quality of document model contributes to the effectiveness of document classification. Conventionally, model evaluation is accomplished by comparing the effectiveness scores of classifiers on model candidates. However, this kind of evaluation methods may encounter either under-fitting or over-fitting problems, because the effectiveness scores are restricted by the learning capacities of classifiers. We propose a model fitness evaluation method to determine whether a model is sufficient to distinguish positive and negative instances while still competent to provide satisfactory effectiveness with a small feature subset. Our experiments demonstrated how the fitness of models are assessed. The results of our work contribute to the researches of feature selection, dimensionality reduction and document classification.

Matching of catalogues by probabilistic pattern classification

We consider the statistical problem of catalogue matching from a machine learning perspective with the goal of producing probabilistic outputs, and using all available information. A framework is provided that unifies two existing approaches to producing probabilistic outputs in the literature, one based on combining distribution estimates and the other based on combining probabilistic classifiers. We apply both of these to the problem of matching the HI Parkes All Sky Survey radio catalogue with large positional uncertainties to the much denser SuperCOSMOS catalogue with much smaller positional uncertainties. We demonstrate the utility of probabilistic outputs by a controllable completeness and efficiency trade-off and by identifying objects that have high probability of being rare. Finally, possible biasing effects in the output of these classifiers are also highlighted and discussed.

Inhibition, encoding specificity, and response availability revisited

In the think/no-think paradigm people practice “suppressing” a learned response to a cue. Practice at suppression appears to produce a long-lasting inhibition of the suppressed response, as evidenced by a subsequent failure to recall the response to an extralist (associatively related, non-studied) cue. Critical to this interpretation is the assumption that suppression practice is necessary. A series of interference paradigms, which do not involve suppression practice and which are structurally similar to the think/no-think paradigm, provide evidence against the inhibition interpretation. Additional evidence against inhibition derives from our demonstrations herewith that the findings from the think/no-think paradigm can be replicated without any apparent suppression requirement. Furthermore, the results from all of these paradigms can be explained by the same simple principle. Namely, that when an item exists in an extended associative network, strengthening the item makes it interfere with the recall of other items in the network.

Boosting the tree augmented naive bayes classifier

The Tree Augmented Naïve Bayes (TAN) classifier relaxes the sweeping independence assumptions of the Naïve Bayes approach by taking account of conditional probabilities. It does this in a limited sense, by incorporating the conditional probability of each attribute given the class and (at most) one other attribute. The method of boosting has previously proven very effective in improving the performance of Naïve Bayes classifiers and in this paper, we investigate its effectiveness on application to the TAN classifier.

Improving support vector solutions by selecting a sequence of training subsets

In this paper we demonstrate that it is possible to gradually improve the performance of support vector machine (SVM) classifiers by using a genetic algorithm to select a sequence of training subsets from the available data. Performance improvement is possible because the SVM solution generally lies some distance away from the Bayes optimal in the space of learning parameters. We illustrate performance improvements on a number of benchmark data sets.

Effectiveness of document representation for classification

Conventionally, document classification researches focus on improving the learning capabilities of classifiers. Nevertheless, according to our observation, the effectiveness of classification is limited by the suitability of document representation. Intuitively, the more features that are used in representation, the more comprehensive that documents are represented. However, if a representation contains too many irrelevant features, the classifier would suffer from not only the curse of high dimensionality, but also overfitting. To address this problem of suitableness of document representations, we present a classifier-independent approach to measure the effectiveness of document representations. Our approach utilises a labelled document corpus to estimate the distribution of documents in the feature space. By looking through documents in this way, we can clearly identify the contributions made by different features toward the document classification. Some experiments have been performed to show how the effectiveness is evaluated. Our approach can be used as a tool to assist feature selection, dimensionality reduction and document classification.

ASSOCIATIVISMO RELIGIOSO E OS IMIGRANTES AFRICANOS: O caso do Campo Missionário Congoangolano da Assembleia de Deus na cidade do Rio de Janeiro

Esta pesquisa faz uma análise do Campo Missionário Congoangolano da Assembleia de Deus, localizado no bairro de Brás de Pina - Zona norte da cidade do Rio de Janeiro. Procura identificar a função desta comunidade religiosa para os imigrantes congoleses e angolanos que a ela pertencem. Desse modo, visa refletir sobre a formação de um espaço territorial religioso consolidado por elementos da religiosidade africana e do pentecostalismo assembleiano e sua imbricada associação com a formação de redes de apoio e de coesão social em torno da manutenção e sustentação de um espaço identitário. Esse espaço é marcado por elementos que expressam símbolos e signos dos países de origem de seus integrantes - Congo e Angola - ao utilizarem a liturgia africana em seus cultos. A pesquisa leva em consideração as demandas que norteiam o processo migratório, as leis que regem esses imigrantes e o quanto tal processo contribui para práticas associativas que envolvem fatores inerentes a inserção e integração sociocultural e econômica no interior do campo missionário.

Religião, redes sociais e capital social no Município de Rio Grande da Serra na Região do Grande ABC Paulista. Estudo das trocas materiais e simbólicas em contexto de periferia urbana

O município de Rio Grande da Serra está situado em uma região Grande ABC paulista reconhecida nacionalmente por seu desenvolvimento econômico e industrial e pelas lutas políticas e sindicais. Paradoxalmente, se configura, social e territorialmente falando, por uma região de periferia urbana. Resultado da forma como a urbanização, na sociedade moderna, conforma o espaço em regiões centrais e periféricas. Localizado no caminho que ligava Santos à Mogi das Cruzes (século XIX) povoado de Geribatiba decorrente das transformações urbanas ocorridas em toda a região, conquistaria, nos anos 1960, sua autonomia político-administrativa. Nas décadas seguintes testemunhou intenso crescimento populacional, resultado do processo migratório, principalmente de mineiros e nordestinos que tinham as cidades, e indústrias, de São Paulo e Grande ABC como destino. Esse deslocamento de pessoas, e as redes formadas em seu em torno, contribuiu para o desenvolvimento de seu campo religioso. Atualmente, com uma população, em torno, de 46 mil habitantes, possui aproximadamente 180 locais de cerimônias religiosas. Nesse contexto, a tese analisa a inserção regional socioeconômica e religiosa de Rio Grande da Serra, a partir de dados comparativos com os demais municípios, e discute como o regionalismo tem contribuído para seu desenvolvimento econômico. Realiza a caracterização das periferias urbanas, discutindo aspectos que lhes são inerentes, como segregação e vulnerabilidade social. Nesse sentido, a investigação possibilitou a identificação do perfil socioeconômico (renda e escolaridade) dos participantes dos grupos religiosos (católicos, evangélicos, kardecistas e umbandistas), permitindo, também, identificar desigualdades sociais no interior de seu território, constatando que determinados bairros são mais vulneráveis do que outros. Considerando que esse estudo examina a capacidade das redes sociais e religiosas, de aumentar o capital social de seus participantes, foi realizado o mapeamento e etnografia das diversas práticas associativas, mais ou menos formais e estruturadas, de forma a analisar os elementos materiais e simbólicos por elas produzidos. Constatou-se, apoiado na aplicação de questionários, entrevistas e observação participativa, que, a partir do habitus religioso de cada grupo, as redes possibilitam no âmbito econômico questões como emprego e renda ou auxílio em necessidades básicas de sobrevivência, através de campanhas e trabalhos sociais. No âmbito simbólico, as redes propiciam questões importantes à existência humana, como a crença na salvação ou evolução da alma, socialização, autoestima, prestígio ou ainda a expectativa de cura ou tratamento de dependência química. Pôde-se aferir que, a despeito das diferentes formas como cada grupo, e seus participantes, se apropriam do capital social, as redes sociais e religiosas, no município, funcionam como redes de proteção, especialmente à população em situação alta de vulnerabilidade social.

RELIGIÃO E CAPITAL SOCIAL NA PERIFERIA URBANA DO GRANDE ABC PAULISTA UMA ANÁLISE DAS REDES SOCIAIS PENTECOSTAIS NO MUNICÍPIO DE RIO GRANDE DA SERRA

A presente pesquisa analisa igrejas evangélicas pentecostais no município de Rio Grande da Serra, região do Grande ABC Paulista. Busca identificar qual é o papel social desses grupos religiosos de grande crescimento no município junto aos seus membros em situação socioeconômica frágil, tanto quanto para a população carente da cidade. A partir de uma reflexão acerca da constituição de espaços territoriais periféricos , e das práticas associativas nesses espaços, nosso objetivo é estudar o associativismo religioso formal e informal, e a construção de redes sociais em seu entorno, e a sua contribuição para o aumento do capital social, entendido este, segundo Bourdieu, como agregado de benefícios materiais e simbólicos, de seus participantes. A pesquisa leva em consideração que a cidade caracteriza-se por ser uma região de periferia, em que parcela importante da população vive em situação de segregação, riscos e vulnerabilidade social.

Blood type classification using computer vision and machine learning

In emergency situations, where time for blood transfusion is reduced, the O negative blood type (the universal donor) is administrated. However, sometimes even the universal donor can cause transfusion reactions that can be fatal to the patient. As commercial systems do not allow fast results and are not suitable for emergency situations, this paper presents the steps considered for the development and validation of a prototype, able to determine blood type compatibilities, even in emergency situations. Thus it is possible, using the developed system, to administer a compatible blood type, since the first blood unit transfused. In order to increase the system’s reliability, this prototype uses different approaches to classify blood types, the first of which is based on Decision Trees and the second one based on support vector machines. The features used to evaluate these classifiers are the standard deviation values, histogram, Histogram of Oriented Gradients and fast Fourier transform, computed on different regions of interest. The main characteristics of the presented prototype are small size, lightweight, easy transportation, ease of use, fast results, high reliability and low cost. These features are perfectly suited for emergency scenarios, where the prototype is expected to be used.

Novel formulation strategies for the fabrication of lyophilised orally disintegrating tablets

Orally disintegrating Tablets (ODTs), also known as fast-disintegrating, fast-melt or fast-dissolving tablets, are a relatively novel dosage technology that involves the rapid disintegration or dissolution of the dosage form into a solution or suspension in the mouth without the need for water. The solution containing the active ingredients is swallowed, and the active ingredients are then absorbed through the gastrointestinal epithelium to reach the target and produce the desired effect. Formulation of ODTs was originally developed to address swallowing difficulties of conventional solid oral dosage forms (tablets and capsules) experienced by wide range of patient population, especially children and elderly. The current work investigates the formulation and development of ODTs prepared by freeze drying. Initial studies focused on formulation parameters that influence the manufacturing process and performance of lyophilised tablets based on excipients used in commercial products (gelatin and saccharides). The second phase of the work was followed up by comprehensive studies to address the essential need to create saccharide free ODTs using naturally accruing amino acids individually or in combinations. Furthermore, a factorial design study was carried out to investigate the feasibility of delivering multiparticulate systems of challenging drugs using a novel formulation that exploited the electrostatic associative interaction between gelatin and carrageenan. Finally, studies aimed to replace gelatin with ethically and morally accepted components to the end users were performed and the selected binder was used in factorial design studies to investigate and optimise ODT formulations that incorporated drugs with varies physicochemical properties. Our results show that formulation of elegant lyophilised ODTs with instant disintegration and adequate mechanical strength requires carful optimisation of gelatin concentration and bloom strength in addition to saccharide type and concentration. Successful formulation of saccharides free lyophilised ODTs requires amino acids that crystallise in the frozen state or display relatively high Tg', interact and integrate completely with the binder and, also, display short wetting time with the disintegrating medium. The use of an optimised mixture of gelatin, carrageenan and alanine was able to create viscous solutions to suspend multiparticulate systems and at the same time provide tablets with short disintegration times and adequate mechanical properties. On the other hand, gum arabic showed an outstanding potential for use as a binder in the formulation of lyophilised ODTs. Compared to gelatin formulations, the use of gum arabic simplified the formulation stages, shortened the freeze drying cycles and produced tablets with superior performance in terms of the disintegration time and mechanical strength. Furthermore, formulation of lyophilised ODTs based on gum arabic showed capability to deliver diverse range of drugs with advantages over commercial products.

Classification of oximetry signals using Bayesian neural networks to assist in the detection of obstructive sleep apnoea syndrome

In the present study, multilayer perceptron (MLP) neural networks were applied to help in the diagnosis of obstructive sleep apnoea syndrome (OSAS). Oxygen saturation (SaO2) recordings from nocturnal pulse oximetry were used for this purpose. We performed time and spectral analysis of these signals to extract 14 features related to OSAS. The performance of two different MLP classifiers was compared: maximum likelihood (ML) and Bayesian (BY) MLP networks. A total of 187 subjects suspected of suffering from OSAS took part in the study. Their SaO2 signals were divided into a training set with 74 recordings and a test set with 113 recordings. BY-MLP networks achieved the best performance on the test set with 85.58% accuracy (87.76% sensitivity and 82.39% specificity). These results were substantially better than those provided by ML-MLP networks, which were affected by overfitting and achieved an accuracy of 76.81% (86.42% sensitivity and 62.83% specificity). Our results suggest that the Bayesian framework is preferred to implement our MLP classifiers. The proposed BY-MLP networks could be used for early OSAS detection. They could contribute to overcome the difficulties of nocturnal polysomnography (PSG) and thus reduce the demand for these studies.

Gaussian processes and SVM: Mean field results and leave-one-out

In this chapter, we elaborate on the well-known relationship between Gaussian processes (GP) and Support Vector Machines (SVM). Secondly, we present approximate solutions for two computational problems arising in GP and SVM. The first one is the calculation of the posterior mean for GP classifiers using a `naive' mean field approach. The second one is a leave-one-out estimator for the generalization error of SVM based on a linear response method. Simulation results on a benchmark dataset show similar performances for the GP mean field algorithm and the SVM algorithm. The approximate leave-one-out estimator is found to be in very good agreement with the exact leave-one-out error.

The TAP approach to intensive and extensive connectivity systems

The Thouless-Anderson-Palmer (TAP) approach was originally developed for analysing the Sherrington-Kirkpatrick model in the study of spin glass models and has been employed since then mainly in the context of extensively connected systems whereby each dynamical variable interacts weakly with the others. Recently, we extended this method for handling general intensively connected systems where each variable has only O(1) connections characterised by strong couplings. However, the new formulation looks quite different with respect to existing analyses and it is only natural to question whether it actually reproduces known results for systems of extensive connectivity. In this chapter, we apply our formulation of the TAP approach to an extensively connected system, the Hopfield associative memory model, showing that it produces identical results to those obtained by the conventional formulation.

Mapping dissociations in verb morphology

Substantial behavioural and neuropsychological evidence has been amassed to support the dual-route model of morphological processing, which distinguishes between a rule-based system for regular items (walk–walked, call–called) and an associative system for the irregular items (go–went). Some neural-network models attempt to explain the neuropsychological and brain-mapping dissociations in terms of single-system associative processing. We show that there are problems in the accounts of homogeneous networks in the light of recent brain-mapping evidence of systematic double-dissociation. We also examine the superior capabilities of more internally differentiated connectionist models, which, under certain conditions, display systematic double-dissociations. It appears that the more differentiation models show, the more easily they account for dissociation patterns, yet without implementing symbolic computations.