Impedance-based structural health monitoring with artificial neural networks

This paper presents a non-model based technique to detect, locate, and characterize structural damage by combining the impedance-based structural health monitoring technique with an artificial neural network. The impedance-based structural health monitoring technique, which utilizes the electromechanical coupling property of piezoelectric materials, has shown engineering feasibility in a variety of practical field applications. Relying on high frequency structural excitations (typically >30 kHz), this technique is very sensitive to minor structural changes in the near field of the piezoelectric sensors. In order to quantitatively assess the state of structures, multiple sets of artificial neural networks, which utilize measured electrical impedance signals for input patterns, were developed. By employing high frequency ranges and by incorporating neural network features, this technique is able to detect the damage in its early stage and to estimate the nature of damage without prior knowledge of the model of structures. The paper concludes with experimental examples, investigations on a massive quarter scale model of a steel bridge section and a space truss structure, in order to verify the performance of this proposed methodology.

Matching Recall and Storage in Sequence Learning with Spiking Neural Networks

Storing and recalling spiking sequences is a general problem the brain needs to solve. It is, however, unclear what type of biologically plausible learning rule is suited to learn a wide class of spatiotemporal activity patterns in a robust way. Here we consider a recurrent network of stochastic spiking neurons composed of both visible and hidden neurons. We derive a generic learning rule that is matched to the neural dynamics by minimizing an upper bound on the Kullback–Leibler divergence from the target distribution to the model distribution. The derived learning rule is consistent with spike-timing dependent plasticity in that a presynaptic spike preceding a postsynaptic spike elicits potentiation while otherwise depression emerges. Furthermore, the learning rule for synapses that target visible neurons can be matched to the recently proposed voltage-triplet rule. The learning rule for synapses that target hidden neurons is modulated by a global factor, which shares properties with astrocytes and gives rise to testable predictions.

Data Mining with Enhanced Neural Networks-CMMSE

Abstract This paper presents a new method to extract knowledge from existing data sets, that is, to extract symbolic rules using the weights of an Artificial Neural Network. The method has been applied to a neural network with special architecture named Enhanced Neural Network (ENN). This architecture improves the results that have been obtained with multilayer perceptron (MLP). The relationship among the knowledge stored in the weights, the performance of the network and the new implemented algorithm to acquire rules from the weights is explained. The method itself gives a model to follow in the knowledge acquisition with ENN.

Efficiency analysis of information technology and online social networks management : an integrated DEA-Model assessment

This paper analyses the relationship between productive efficiency and online-social-networks (OSN) in Spanish telecommunications firms. A data-envelopment-analysis (DEA) is used and several indicators of business ?social Media? activities are incorporated. A super-efficiency analysis and bootstrapping techniques are performed to increase the model?s robustness and accuracy. Then, a logistic regression model is applied to characterise factors and drivers of good performance in OSN. Results reveal the company?s ability to absorb and utilise OSNs as a key factor in improving the productive efficiency. This paper presents a model for assessing the strategic performance of the presence and activity in OSN.

Regression models with MoPs Bayesian networks

We present a model of Bayesian network for continuous variables, where densities and conditional densities are estimated with B-spline MoPs. We use a novel approach to directly obtain conditional densities estimation using B-spline properties. In particular we implement naive Bayes and wrapper variables selection. Finally we apply our techniques to the problem of predicting neurons morphological variables from electrophysiological ones.

Computation with infinite neural networks

For neural networks with a wide class of weight priors, it can be shown that in the limit of an infinite number of hidden units, the prior over functions tends to a gaussian process. In this article, analytic forms are derived for the covariance function of the gaussian processes corresponding to networks with sigmoidal and gaussian hidden units. This allows predictions to be made efficiently using networks with an infinite number of hidden units and shows, somewhat paradoxically, that it may be easier to carry out Bayesian prediction with infinite networks rather than finite ones.

Shared Education in contested spaces: How collaborative networks improve communities and schools

Societies which suffer from ethnic and political divisions are often characterised by patterns of social and institutional separation, and sometimes these divisions remain even after political conflict has ended. This has occurred in Northern Ireland where there is, and remains, a long-standing pattern of parallel institutions and services for the different communities. A socially significant example lies in the education system where a parallel system of Catholic and Protestant schools has been in place since the establishment of a national school system in the 1830s. During the years of political violence in Northern Ireland a variety of educational interventions were implemented to promote reconciliation, but most of them failed to create any systemic change. This paper describes a post-conflict educational initiative known as Shared Education which aims to promote social cohesion and school improvement by encouraging sustained and regular shared learning between students and broader collaboration between teachers and school leaders from different schools. The paper examines the background to work on Shared Education, describes a ‘sharing continuum’ which emerged as an evaluation and policy tool from this work and considers evidence from a case study of a Shared Education school partnership in a divided city in Northern Ireland. The paper will conclude by highlighting some of the significant social and policy impact of the Shared Education work.

Adverse drug reaction classification with deep neural networks

We study the problem of detecting sentences describing adverse drug reactions (ADRs) and frame the problem as binary classification. We investigate different neural network (NN) architectures for ADR classification. In particular, we propose two new neural network models, Convolutional Recurrent Neural Network (CRNN) by concatenating convolutional neural networks with recurrent neural networks, and Convolutional Neural Network with Attention (CNNA) by adding attention weights into convolutional neural networks. We evaluate various NN architectures on a Twitter dataset containing informal language and an Adverse Drug Effects (ADE) dataset constructed by sampling from MEDLINE case reports. Experimental results show that all the NN architectures outperform the traditional maximum entropy classifiers trained from n-grams with different weighting strategies considerably on both datasets. On the Twitter dataset, all the NN architectures perform similarly. But on the ADE dataset, CNN performs better than other more complex CNN variants. Nevertheless, CNNA allows the visualisation of attention weights of words when making classification decisions and hence is more appropriate for the extraction of word subsequences describing ADRs.

Design and use of mobile technology in distance language education : Matching learning practices with technologies-in-practice

This thesis focuses on the adaptation of formal education to people’s technology- use patterns, theirtechnology-in-practice, where the ubiquitous use of mobile technologies is central. The research question is: How can language learning practices occuring in informal learning environments be effectively integrated with formal education through the use of mobile technology? The study investigates the technical, pedagogical, social and cultural challenges involved in a design science approach. The thesis consists of four studies. The first study systematises MALL (mobile-assisted language learning) research. The second investigates Swedish and Chinese students’ attitudes towards the use of mobile technology in education. The third examines students’ use of technology in an online language course, with a specific focus on their learning practices in informal learning contexts and their understanding of how this use guides their learning. Based on the findings, a specifically designed MALL application was built and used in two courses. Study four analyses the app use in terms of students’ perceived level of self-regulation and structuration. The studies show that technology itself plays a very important role in reshaping peoples’ attitudes and that new learning methods are coconstructed in a sociotechnical system. Technology’s influence on student practices is equally strong across borders. Students’ established technologies-in-practice guide the ways they approach learning. Hence, designing effective online distance education involves three interrelated elements: technology, information, and social arrangements. This thesis contributes to mobile learning research by offering empirically and theoretically grounded insights that shift the focus from technology design to design of information systems.

Sentidos da docência para egressos dos cursos de licenciatura a distância da Universidade Federal do Espírito Santo no Polo de Itapemirim – Espírito Santo

Investiga os sentidos do ser e dos saberes docentes com ênfase no processo de formação inicial de professores para a educação básica, por meio da modalidade de Educação a Distância – EAD. Questiona os modos como o ser docente e os saberes da profissão docente foram se constituindo ao longo das trajetórias de formação percorridas pelos estudantes egressos dos três primeiros cursos de licenciatura em Química, Física e Artes Visuais, ofertados pela Universidade Federal do Espírito Santo (Ufes), no âmbito do Sistema Universidade Aberta do Brasil – UAB, entre 2008 e 2014, no Polo da cidade de Itapemirim/ES, em um recorte temporal definido como antes, durante e após a integralização dos referidos cursos. Pressupõe uma perspectiva teórica crítica, que compreende a formação e a docência como processos históricos de construção social e coletiva, que não possuem início e término em si, por si e para si. A pesquisa delineia-se como um estudo de caso qualitativo. A abordagem aos sujeitos deu-se por meio de técnicas que visam à coleta de dados descritivos, com o uso de três principais instrumentos, no formato semiestruturado: um questionário, uma entrevista coletiva e um fórum virtual temático. Para compor o repertório de dados, ocorreram, ainda, informações advindas dos documentos e bases legais que deram sustentação à oferta dos cursos, bem como dos relatórios de acompanhamento e gestão destes. A análise dos dados se deu por meio da técnica de triangulação, com sustentação teórica nos estudos de Freire, Nóvoa e Tardif. Evidencia a necessidade de estudos na área da formação articulada à EAD, bem como da consideração dos saberes cotidianos da docência na proposição de políticas à sua formação. Revela uma variedade de sentidos atribuídos aos conceitos de docência e dos saberes da docência e a sua constituição em meio a processos formativos ao longo de toda a vida dos sujeitos. Reconhece a necessidade de uma formação contínua do docente após a obtenção de sua titulação profissional e aponta a EAD como possibilidade de acesso a essa formação na/pela Universidade pública, em tempos e espaços que se vêm constituindo, em meio às novas tecnologias da informação e da comunicação. Ressalta a importância do Sistema UAB para a disseminação da formação e da EAD, bem como a necessidade de institucionalização dessa modalidade como forma de superação do caráter emergencial e provisório da atual política de formação de professores no Brasil.

Modelos e interfaces gráficos para objectos de aprendizagem na área da saúde

A educação é uma área bastante importante no desenvolvimento humano e tem vindo a adaptar-se às novas tecnologias. Tentam-se encontrar novas maneiras de ensinar de modo a obter um rendimento cada vez maior na aprendizagem das pessoas. Com o aparecimento de novas tecnologias como os computadores e a Internet, a concepção de aplicações digitais educativas cresceu e a necessidade de instruir cada vez melhor os alunos leva a que estas aplicações precisem de um interface que consiga leccionar de uma maneira rápida e eficiente. A combinação entre o ensino com o auxílio dessas novas tecnologias e a educação à distância deu origem ao e-Learning (ensino à distância). Através do ensino à distância, as possibilidades de aumento de conhecimento dos alunos aumentaram e a informação necessária tornou-se disponível a qualquer hora em qualquer lugar com acesso à Internet. Mas os cursos criados online tinham custos altos e levavam muito tempo a preparar o que gerou um problema para quem os criava. Para recuperar o investimento realizado decidiu-se dividir os conteúdos em módulos capazes de serem reaproveitados em diferentes contextos e diferentes tipos de utilizadores. Estes conteúdos modulares foram denominados Objectos de Aprendizagem. Nesta tese, é abordado o estudo dos Objectos de Aprendizagem e a sua evolução ao longo dos tempos em termos de interface com o utilizador. A concepção de um interface que seja natural e simples de utilizar nem sempre é fácil e independentemente do contexto em que se insere, requer algum conhecimento de regras que façam com que o utilizador que use determinada aplicação consiga trabalhar com um mínimo de desempenho. Na concepção de Objectos de Aprendizagem, áreas de complexidade elevada como a Medicina levam a que professores ou doutores sintam alguma dificuldade em criar um interface com conteúdos educativos capaz de ensinar com eficiência os alunos, devido ao facto de grande parte deles desconhecerem as técnicas e regras que levam ao desenvolvimento de um interface de uma aplicação. Através do estudo dessas regras e estilos de interacção torna-se mais fácil a criação de um bom interface e ao longo desta tese será estudado e proposto uma ferramenta que ajude tanto na criação de Objectos de Aprendizagem como na concepção do respectivo interface.

Learning networks and moodle use in online courses: a social network analysis study

Dissertação para obtenção do Grau de Doutor em Ciências da Educação Especialidade em Tecnologias, Redes e Multimédia na Educação e Formação

Adaptive intelligent agent approach to guide the web navigation on the PLAN-G distance learning platform

Hypermedia systems based on the Web for open distance education are becoming increasinglypopular as tools for user-driven access learning information. Adaptive hypermedia is a new direction in research within the area of user-adaptive systems, to increase its functionality by making it personalized [Eklu 961. This paper sketches a general agents architecture to include navigationaladaptability and user-friendly processes which would guide and accompany the student during hislher learning on the PLAN-G hypermedia system (New Generation Telematics Platform to Support Open and Distance Learning), with the aid of computer networks and specifically WWW technology [Marz 98-1] [Marz 98-2]. The PLAN-G actual prototype is successfully used with some informatics courses (the current version has no agents yet). The propased multi-agent system, contains two different types of adaptive autonomous software agents: Personal Digital Agents {Interface), to interacl directly with the student when necessary; and Information Agents (Intermediaries), to filtrate and discover information to learn and to adapt navigation space to a specific student