Learning is one such innate general cognitive ability which has empowered the living animate entities and especially humans with intelligence. It is obtained by acquiring new knowledge and skills that enable them to adapt and survive. With the advancement of technology, a large amount of information gets amassed. Due to the sheer volume of increasing information, its analysis is humanly unfeasible and impractical. Therefore, for the analysis of massive data we need machines (such as computers) with the ability to learn and evolve in order to discover new knowledge from the analysed data. The majority of the traditional machine learning algorithms function optimally on a parametric (static) data. However, the datasets acquired in real practices are often vast, inaccurate, inconsistent, non-parametric and highly volatile. Therefore, the learning algorithms’ optimized performance can only be transitory, thus requiring a learning algorithm that can constantly evolve and adapt according to the data it processes. In light of a need for such machine learning algorithm, we look for the inspiration in humans’ innate cognitive learning ability. Active learning is one such biologically inspired model, designed to mimic humans’ dynamic, evolving, adaptive and intelligent cognitive learning ability. Active learning is a class of learning algorithms that aim to create an accurate classifier by iteratively selecting essentially important unlabeled data points by the means of adaptive querying and training the classifier on those data points which are potentially useful for the targeted learning task (Tong & Koller, 2002). The traditional active learning techniques are implemented under supervised or semi-supervised learning settings (Pang et al., 2009). Our proposed model performs the active learning in an unsupervised setting by introducing a discriminative selective sampling criterion, which reduces the computational cost by substantially decreasing the number of irrelevant instances to be learned by the classifier. The methods based on passive learning (which assumes the entire dataset for training is truly informative and is presented in advance) prove to be inadequate in a real world application (Pang et al., 2009). To overcome this limitation, we have developed Active Mode Incremental Nonparametric Discriminant Analysis (aIncNDA) which undertakes adaptive discriminant selection of the instances for an incremental NDA learning. NDA is a discriminant analysis method that has been incorporated in our selective sampling technique in order to reduce the effects of the outliers (which are anomalous observations/data points in a dataset). It works with significant efficiency on the anomalous datasets, thereby minimizing the computational cost (Raducanu & Vitri´a, 2008). NDA is one of the methods used in the proposed active learning model. This thesis presents the research on a discrimination-based active learning where NDA is extended for fast discrimination analysis and data sampling. In addition to NDA, a base classifier (such as Support Vector Machine (SVM) and k-Nearest Neighbor (k-NN)) is applied to discover and merge the knowledge from the newly acquired data. The performance of our proposed method is evaluated against benchmark University of California, Irvine (UCI) datasets, face image, and object image category datasets. The assessment that was carried out on the UCI datasets showed that Active Mode Incremental NDA (aIncNDA) performs at par and in many cases better than the incremental NDA with a lower number of instances. Additionally, aIncNDA also performs efficiently under the different levels of redundancy, but has an improved discrimination performance more often than a passive incremental NDA. In an application that undertakes the face image and object image recognition and retrieval task, it can be seen that the proposed multi-example active learning system dynamically and incrementally learns from the newly obtained images, thereby gradually reducing its retrieval (classification) error rate by the means of iterative refinement. The results of the empirical investigation show that our proposed active learning model can be used for classification with increased efficiency. Furthermore, given the nature of network data which is large, streaming, and constantly changing, we believe that our method can find practical application in the field of Internet security.

Thesis (M.S.)--Brigham Young University. Dept. of Civil and Environmental Engineering, 2006. / Includes bibliographical references (p. 91).

Orientador: Fábio Santos de Lira / Banca: Manoel Carlos Spiguel Lima / Banca: Valeria Leme Gonçalves Panissa / Resumo: Introdução: A dor lombar crônica (DLC) não especifica é a mais encontrada na prática clínica, além de ser um dos problemas mais discutidos no mundo. Indivíduos com dor lombar crônica (CDLC) não especifica possuem um menor nível de atividade física quando comparados com indivíduos sem dor lombar crônica (SDLC), e, portanto, uma diminuiição na potência aeróbia (VO2máx). As diretrizes da área de dor lombar recomendam a prática de exercício físico, assim como afirmam ser eficaz em diminuir a intensidade da dor e incapacidade funcional. No entanto, os estudos que utilizam o exercício físico como tratamento para DLC, falham ao não prescrever o exercício de maneira específica ... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Introduction: The non-specific chronic low back pain (NSCLBP) is the most commonly found in clinical practice as well as being one of the most discussed problems in the world. Individuals with non-specific chronic low back pain (NSCLBP) have a lower level of physical activity when compared with individuals without NSCLBP, and therefore a decrease in aerobic power (V̇O2max). The guidelines of NSCLBP recommend as treatment the practice of physical exercise, as they claim to be effective in reducing the intensity of pain and functional disability. However, studies that use exercise as a treatment for LBP fail to prescribe exercise in a specific way. Objective: Evaluate and compare the effectiveness of different treadmill protocols for aerobic fitness determination in individuals with and without NSCLBP. Materials and Methods: Twenty individuals (10 with NSCLBP and 10 without NSCLBP) underwent two tests (modified Bruce and incremental) on treadmill. The following were collected during the tests: oxygen consumption (V̇O2), heart rate (HR), rate perceived exertion (RPE 6-20), blood lactate concentration ([Lac]), respiratory quotient (RQ) and pain intensity. Independent t-test was used to evaluate the difference between groups. Differences in physiological responses within groups (with and without NSCLBP) and between groups (Group vs. Test Interaction) were analyzed using two-way ANOVA. Statistical significance was set at p < 0.05 ... (Complete abstract click electronic access below) / Mestre

Este trabalho descreve a adaptação de um equipamento CNC e ferramentas computacionais CAD/CAM (desenvolvidas originalmente para a área de usinagem) na conformação de chapas de alumínio comercialmente puro, através do processo de estampagem incremental. Através das simulações e dos testes reais foi possível obter resultados práticos, relativos às limitações geométricas e ao acabamento superficial. As peças conformadas têm o formato básico de um tronco de cone, onde foram alterados a profundidade e o ângulo do cone. A análise dimensional das peças, comparadas com seu respectivo modelo CAD revelou razoável conformidade geométrica. Por outro lado, ângulos de parede maiores que 65° causaram o rompimento da chapa. As estampagens foram realizadas em um centro de usinagem, e utilizaram-se como valores máximos a velocidade de avanço de 2.200 mm/min e o incremento vertical de 1mm, por questões de segurança e também pela constatação de problemas de acabamento superficial e de vibrações excessivas do equipamento. Como conclusão final, foi demonstrado que, para o tipo específico de peça (material, espessura e geometria), dispositivo prensa-chapas, ferramentas, equipamento CNC e softwares utilizados, o processo de estampagem incremental seria viável para lotes de até 600 peças. / This paper describes the adaptation of a CNC equipment and CAD/CAM computational tools (originally developed for the machining area) on the conformation of sheets of commercially pure aluminum, through the process of incremental forming. Through simulations and tests were unable to get practical results, for the geometric constraints and surface finishing. Forming parts have the basic format of a truncated cone, which have changed the depth and angle of the cone. A dimensional analysis of the parts, compared with their CAD model showed geometric reasonable compliance. Moreover, the wall angles greater than 65 ° caused the breaking of the plate. The sheets formings were performed in a machining center, with a maximum speed of advance of 2,200 mm/min and the vertical increase of 1mm, for reasons of safety and also for finding finishing surface problems and the excessive vibration equipment. As final conclusion, it was demonstrated that, for the specific type of piece (material, thickness and geometry), device-press plates, tools, CNC equipment and software used, the process of incremental sheet forming feasible for batches of up to 600 pieces.

Many problems in nature can be represented as some kind of a satisfiability problem. Several SAT solvers and SMT solvers have been developed in the last decade with the goal of checking the satisfiability of different SAT problems. An all-solution satisfiability modulo theories on top of the Z3 SMT solver is presented that uses the clause blocking algorithm to find all the solution sets of a SAT problem. Then, an incremental All-SMT solver has been presented based on the all-SMT solver which is able to find the satisfiable answers of an incremental SMT problem based on the solution set of the original problem.

Este trabalho descreve a adaptação de um equipamento CNC e ferramentas computacionais CAD/CAM (desenvolvidas originalmente para a área de usinagem) na conformação de chapas de alumínio comercialmente puro, através do processo de estampagem incremental. Através das simulações e dos testes reais foi possível obter resultados práticos, relativos às limitações geométricas e ao acabamento superficial. As peças conformadas têm o formato básico de um tronco de cone, onde foram alterados a profundidade e o ângulo do cone. A análise dimensional das peças, comparadas com seu respectivo modelo CAD revelou razoável conformidade geométrica. Por outro lado, ângulos de parede maiores que 65° causaram o rompimento da chapa. As estampagens foram realizadas em um centro de usinagem, e utilizaram-se como valores máximos a velocidade de avanço de 2.200 mm/min e o incremento vertical de 1mm, por questões de segurança e também pela constatação de problemas de acabamento superficial e de vibrações excessivas do equipamento. Como conclusão final, foi demonstrado que, para o tipo específico de peça (material, espessura e geometria), dispositivo prensa-chapas, ferramentas, equipamento CNC e softwares utilizados, o processo de estampagem incremental seria viável para lotes de até 600 peças. / This paper describes the adaptation of a CNC equipment and CAD/CAM computational tools (originally developed for the machining area) on the conformation of sheets of commercially pure aluminum, through the process of incremental forming. Through simulations and tests were unable to get practical results, for the geometric constraints and surface finishing. Forming parts have the basic format of a truncated cone, which have changed the depth and angle of the cone. A dimensional analysis of the parts, compared with their CAD model showed geometric reasonable compliance. Moreover, the wall angles greater than 65 ° caused the breaking of the plate. The sheets formings were performed in a machining center, with a maximum speed of advance of 2,200 mm/min and the vertical increase of 1mm, for reasons of safety and also for finding finishing surface problems and the excessive vibration equipment. As final conclusion, it was demonstrated that, for the specific type of piece (material, thickness and geometry), device-press plates, tools, CNC equipment and software used, the process of incremental sheet forming feasible for batches of up to 600 pieces.

Este trabalho descreve a adaptação de um equipamento CNC e ferramentas computacionais CAD/CAM (desenvolvidas originalmente para a área de usinagem) na conformação de chapas de alumínio comercialmente puro, através do processo de estampagem incremental. Através das simulações e dos testes reais foi possível obter resultados práticos, relativos às limitações geométricas e ao acabamento superficial. As peças conformadas têm o formato básico de um tronco de cone, onde foram alterados a profundidade e o ângulo do cone. A análise dimensional das peças, comparadas com seu respectivo modelo CAD revelou razoável conformidade geométrica. Por outro lado, ângulos de parede maiores que 65° causaram o rompimento da chapa. As estampagens foram realizadas em um centro de usinagem, e utilizaram-se como valores máximos a velocidade de avanço de 2.200 mm/min e o incremento vertical de 1mm, por questões de segurança e também pela constatação de problemas de acabamento superficial e de vibrações excessivas do equipamento. Como conclusão final, foi demonstrado que, para o tipo específico de peça (material, espessura e geometria), dispositivo prensa-chapas, ferramentas, equipamento CNC e softwares utilizados, o processo de estampagem incremental seria viável para lotes de até 600 peças. / This paper describes the adaptation of a CNC equipment and CAD/CAM computational tools (originally developed for the machining area) on the conformation of sheets of commercially pure aluminum, through the process of incremental forming. Through simulations and tests were unable to get practical results, for the geometric constraints and surface finishing. Forming parts have the basic format of a truncated cone, which have changed the depth and angle of the cone. A dimensional analysis of the parts, compared with their CAD model showed geometric reasonable compliance. Moreover, the wall angles greater than 65 ° caused the breaking of the plate. The sheets formings were performed in a machining center, with a maximum speed of advance of 2,200 mm/min and the vertical increase of 1mm, for reasons of safety and also for finding finishing surface problems and the excessive vibration equipment. As final conclusion, it was demonstrated that, for the specific type of piece (material, thickness and geometry), device-press plates, tools, CNC equipment and software used, the process of incremental sheet forming feasible for batches of up to 600 pieces.

Ingeniero Civil / El trabajo de Título está acotado por dos objetivos generales, como se desprende del título del tema: en una primera etapa se creará un catálogo de terremotos, el cual almacenará parámetros sísmicos calculados a partir de registros de terremotos relevantes de Chile. Y en una segunda etapa se evaluará la vulnerabilidad de un modelo estructural simplificado con características del tipo de edificio que sufrió mayor daño para el terremoto del Maule de 2010. Para desarrollar la primera etapa del trabajo será necesaria una recopilación de los registros de aceleraciones de las estaciones instaladas en todo el territorio chileno, además de la actualización de las bases de datos ya creadas, que contienen información de eventos, registros y estaciones. A partir de estas bases de datos, se calcularán los parámetros sísmicos más usados (estos son: PGA, PGV, PGD, Intensidad de Arias, Intensidad Espectral de Housner, Potencial Destructivo, Velocidad Absoluta Acumulada, Intensidad JMA) y almacenarán en el catálogo con la información básica de cada evento, tal como, estación, equipo, tipo de suelo y fuente sismogénica. Este catálogo permitirá analizar gráficamente la tendencia entre los distintos parámetros y PGA para todos los registros. Posteriormente, los registros serán escalados, y de esta manera se fijará un rango de PGA en que los valores escalados se ajustan mejor a los no escalados. Con ello, usar el conjunto de registros que representan mejor los parámetros sísmicos de la base de datos de terremotos Subductivos Chilenos para generar las curvas de fragilidad, usando el análisis dinámico incremental (IDA). Dicho análisis escala los registros y permite obtener la respuesta no lineal de un modelo estructural simplificado para los distintos rangos de PGA. Este método se aplicará para el grupo seleccionado de registros y para todo el universo de registros del catálogo, con ello se compararán las curvas y se identificarán las diferencias. De esta manera es posible concluir la influencia de la selección de los registros en análisis dinámico incremental (IDA) y las curvas de fragilidad estudiadas.

This dissertation establishes a novel system for human face learning and recognition based on incremental multilinear Principal Component Analysis (PCA). Most of the existing face recognition systems need training data during the learning process. The system as proposed in this dissertation utilizes an unsupervised or weakly supervised learning approach, in which the learning phase requires a minimal amount of training data. It also overcomes the inability of traditional systems to adapt to the testing phase as the decision process for the newly acquired images continues to rely on that same old training data set. Consequently when a new training set is to be used, the traditional approach will require that the entire eigensystem will have to be generated again. However, as a means to speed up this computational process, the proposed method uses the eigensystem generated from the old training set together with the new images to generate more effectively the new eigensystem in a so-called incremental learning process. In the empirical evaluation phase, there are two key factors that are essential in evaluating the performance of the proposed method: (1) recognition accuracy and (2) computational complexity. In order to establish the most suitable algorithm for this research, a comparative analysis of the best performing methods has been carried out first. The results of the comparative analysis advocated for the initial utilization of the multilinear PCA in our research. As for the consideration of the issue of computational complexity for the subspace update procedure, a novel incremental algorithm, which combines the traditional sequential Karhunen-Loeve (SKL) algorithm with the newly developed incremental modified fast PCA algorithm, was established. In order to utilize the multilinear PCA in the incremental process, a new unfolding method was developed to affix the newly added data at the end of the previous data. The results of the incremental process based on these two methods were obtained to bear out these new theoretical improvements. Some object tracking results using video images are also provided as another challenging task to prove the soundness of this incremental multilinear learning method.

The Caribbean steelpan is a pitched percussion instrument that originated in Trinidad and Tobago during the Second World War. Despite several research initiatives to improve the making of this relatively new instrument, several areas remain unaddressed. This thesis presents new approaches to help improve the making of the instrument. These approaches are situated in the production, vibration and material aspect of the steelpan. A novel sheet forming technology termed Incremental Sheet Forming (ISF) is applied to the production of miniature steelpan dishes. The thickness distribution in the wall of the ISF dishes is compared to the wall thickness distribution in a traditionally formed steelpan dish and a wheeled dish. Unlike traditional forming and wheeling, ISF produces stretching in only a portion of the walls of the formed dishes. Multi-pass ISF is used to extend the stretched zone but this extension is minimal. A break even analysis is also applied to investigate the fiscal viability of ISF application to the production of miniature and full size steelpan dishes. The application of ISF to steelpan making is found to be commercially profitable but could be jeopardised by the tuning stage of the steelpan making process. A preliminary study on the effect of impact on tone stability is conducted on a pair of notes on a fullsize steelpan and detuning is found more likely to occur by repeated impact of the note at its centre. Mode confinement in test-pans is also investigated. ISF is used to produce miniature test-pans with test-notes that are geometrically identical to notes on full size pans. It is possible to confine modes by varying the curvature of the bowl surrounding the test-note. The number of localised modes in the test-note increases as the radius of curvature of the surrounding bowl increases. The natural frequency of the first confined mode in the test-notes is sensitive to material springback in ISF and the mechanism of confinement appears to be due to the change in geometry that occurs between the flat test-note region and the bowl wall. This control of mode confinement may find use in future efforts to completely or partially automate the steelpan making process. Material damping and mechanical properties in low-carbon steel used to produce steelpans are researched. Damping and mechanical properties are extracted from low-carbon steel that is subjected to identical stages to the steelpan production process. Material damping trends suggest that an annealing temperature between 300°C and 400°C would be appropriate for the heat treatment of steelpans. Air-cooled and water-quenched low-carbon specimens exhibit comparable damping trends. Hardness increases in cold formed low-carbon specimens is attributed to strain hardening and not strain ageing. Investigation of damping trends and mechanical properties in ultra low bake-hardenable and interstitial-free steels reveals that a wider range of low-carbon steels may be suitable for steelpan making.