In this thesis work, we propose a Map-Reduce variant of the Growing Hierarchical Self Organizing Map (GHSOM) called MR-GHSOM, which is capable of handling mixed attribute datasets of massive size. The Self Organizing Map (SOM) has proved to be a useful unsupervised data analysis algorithm. It projects a high dimensional data onto a lower dimensional grid of neurons. However, the SOM has some limitations owing to its static structure and the incapability to mirror the hierarchical relations in the data. The GHSOM overcomes these shortcomings of the SOM by providing a dynamic structure that adapts its shape according to the input data. It is capable of growing dynamically in terms of the size of the individual neuron layers to represent data at the desired granularity as well as in depth to model the hierarchical relations in the data. However, the training of the GHSOM requires multiple passes over an input dataset. This makes it difficult to use the GHSOM for massive datasets. In this thesis work, we propose a Map-Reduce variant of the GHSOM called MR-GHSOM, which is capable of processing massive datasets. The MR-GHSOM is implemented using the Apache Spark cluster computing engine and leverages the popular Map-Reduce programming model. This enables us to exploit the usefulness and dynamic capabilities of the GHSOM even for a large dataset. Moreover, the conventional GHSOM algorithm can handle datasets with numeric attributes only. This is owing to the fact that it relies heavily on the Euclidean space dissimilarity measures of the attribute vectors. The MR-GHSOM further extends the GHSOM to handle mixed attribute - numeric and categorical - datasets. It accomplishes this by adopting the distance hierarchy approach of managing mixed attribute datasets. The proposed MR-GHSOM is thus capable of handling massive datasets containing mixed attributes. To demonstrate the effectiveness of the MR-GHSOM in terms of clustering of mixed attribute datasets, we present the results produced by the MR-GHSOM on some popular datasets. We further train our MR-GHSOM on a Census dataset containing mixed attributes and provide an analysis of the results.

Unmanned Aerial Vehicles (UAVs) may develop cracks, erosion, delamination or other damages due to aging, fatigue or extreme loads. Identifying these damages is critical for the safe and reliable operation of the systems. Structural Health Monitoring (SHM) is capable of determining the conditions of systems automatically and continually through processing and interpreting the data collected from a network of sensors embedded into the systems. With the desired awareness of the systems’ health conditions, SHM can greatly reduce operational cost and speed up maintenance processes. The purpose of this study is to develop an effective, low-cost, flexible and fault tolerant structural health monitoring system. The proposed Index Based Reasoning (IBR) system started as a simple look-up-table based diagnostic system. Later, Fast Fourier Transformation analysis and neural network diagnosis with self-learning capabilities were added. The current version is capable of classifying different health conditions with the learned characteristic patterns, after training with the sensory data acquired from the operating system under different status. The proposed IBR systems are hierarchy and distributed networks deployed into systems to monitor their health conditions. Each IBR node processes the sensory data to extract the features of the signal. Classifying tools are then used to evaluate the local conditions with health index (HI) values. The HI values will be carried to other IBR nodes in the next level of the structured network. The overall health condition of the system can be obtained by evaluating all the local health conditions. The performance of IBR systems has been evaluated by both simulation and experimental studies. The IBR system has been proven successful on simulated cases of a turbojet engine, a high displacement actuator, and a quad rotor helicopter. For its application on experimental data of a four rotor helicopter, IBR also performed acceptably accurate. The proposed IBR system is a perfect fit for the low-cost UAVs to be the onboard structural health management system. It can also be a backup system for aircraft and advanced Space Utility Vehicles.

Im Bereich der medizinischen Geräte haben eingebettete Systeme verstärkt Einzug gehalten. Nicht nur in Operationssälen oder Intensivstationen, auch im Bereich der Prothesensteuerung spielt moderne Computertechnik in zunehmendem Maße eine Rolle, auch und insbesondere im Bereich der Prothesensteuerung durch elektronisch vorverarbeitete Nervensignale. Die zur Signalverarbeitung eingesetzte, vortrainierte selbst-organisierende Karte stößt jedoch auf das Problem, sich den verändernden Gegebenheiten in den Nervensignalen des Patienten nicht anpassen zu können. In dieser Arbeit wird die Möglichkeit untersucht, die Steuerung der Handprothese mit einer Nachlernfunktion auszustatten, um während des Einsatzes der Prothese auf die Veränderungen der Nervensignale des Patienten reagieren zu können. Da diese Veränderungen höchst individuell verlaufen, werden Parameter eingeführt, mit denen das Nachlernverfahren an die Gegebenheiten des Patienten angepasst werden kann. Verschiedene denkbare Lernstrategien werden untersucht und hinsichtlich ihrer Effizienz und ihrer Aktualität bewertet. Um die Verwendbarkeit der Implementierung sicherzustellen, muss darauf geachtet werden, dass der entstehende SystemC-Code keine Elemente des nicht synthetisierbaren Subsets enthält. Zusätzlich wird die Synthetisierbarkeit mit dem Agility-Compiler untersucht.

No description available.

Classification tasks are an integral part of science, industry, medicine, and business; being such a pervasive technique, its smallest improvement is valuable. Artificial Neural Network (ANN) is one of the strongest techniques used in many disciplines for classification. The ANN technique suffers from drawbacks such as intransparency in spite of its high prediction power. In this dissertation, motivated by learning styles in human brains, ANN’s shortcomings are assuaged and its learning power is improved. Self-Organizing Map (SOM), an ANN variation which has strong unsupervised power, and Feedforward ANN, traditionally used for classification tasks, are hybridized to solidify their benefits and help remove their limitations. These benefits are in two directions: enhancing ANN’s learning power, and improving decision-making. First, the proposed method, named Self-Organizing Error-Driven (SOED) Artificial Neural Network (ANN), shows significant improvements in comparison with usual ANNs. We show SOED is a more accurate, more reliable, and more transparent technique through experimentation with five famous benchmark datasets. Second, the hybridization creates space for inclusion of decision-making goals at the level of ANN’s learning. This gives the classifier the opportunity to handle the inconclusiveness of the data smarter and in the direction of decision-making goals. Through three case studies, naming 1) churn decision analytics, 2) breast cancer diagnosis, and 3) quality control decision making through thermal monitoring of additive manufacturing processes, this novel and cost-sensitive aspect of SOED has been explored and lead to much quantified improvement in decision-making.

The Fennoscandian Shield as a part of a large Precambrian basement area is located in northern Europe and hosts economically important mineral deposits including base metals and precious metals. Regional geophysical data such as potential field and magnetotelluric data in combination with other geoscientific data contain information of importance for an understanding of the crustal and upper mantle structure. Knowledge about regional-scale structures is important for an optimized search for mineralisation. In order to investigate in more detail the spatial distribution of regional electrically conductive structures and near-surface mineral deposits, complementary magnetotelluric measurements have been done within the Precambrian Shield in the north-eastern part of the Norrbotten ore province. The potential field data provided by the Geological Survey of Sweden have been included in the current study. Processing of magnetotelluric data was performed using a robust multi-remote reference technique. The dimensionality analysis of the phase tensors indicates complex 3D structures in the area. A 3D crustal model of the electrical conductivity structure was derived based on 3D inversion of the data using the ModEM code. The final inversion 3D resistivity model revealed the presence of strong crustal conductors with the conductance of more than 3000 S at depth of tens of kilometres within a generally resistive crust. A significant part of the middle crust conductors is elongated in directions that coincide with major ductile deformation zones that have been mapped from airborne magnetic data and geological fieldwork. Some of these conductors have near-surface expression where they spatially correlate with the locations of known mineralisation. Processing and 3D inversion of the regional magnetic and gravity field data were performed, and the structural information derived from these data by using an open-source object-oriented package code written in Python called SimPEG. In this study, a new approach is proposed to extract and analyse the correlation between the modelled physical properties and for domain classification. For this, a neural net Self-Organizing Map procedure (SOM) was used for data reduction and simplification. The input data to the SOM analysis contain resistivity, magnetic susceptibility, and density model values for some selected depth levels. The domain classification is discussed with respect to geological boundaries and composition. The classification is furthermore applied for prediction of favourable areas for mineralisation. Based on visual inspection of processed regional gravity and magnetic field data and a SOM analysis performed on higher-order derivatives of the magnetic data, an interpretation of a sinistral fault with 52 km offset is proposed. The fault is oriented N10E and can be traced 250 km from Karesuando at the Swedish-Finish border southwards to the Archaean-Proterozoic boundary marked by the Luleå-Jokkmokk Zone.

In this paper, the problem of analyzing hyperspectral data is presented. The complexity of multi-dimensional data leads to the need for computer assisted data compression and labeling of important features. A brief overview of Self-Organizing Maps and their variants is given and then two possible methods of data analysis are examined. These methods are incorporated into a program derived from som_toolbox2. In this program, ASD data (data collected by an Analytical Spectral Device sensor) is read into a variable, relevant bands for discrimination between classes are extracted, and several different methods of analyzing the results are employed. A GUI was developed for easy implementation of these three stages.

Une nouvelle approche du traitement de l'image, appelée SOM-QE, qui exploite quantization error (QE) des self-organizing maps (SOM) est proposée dans cette thèse. Les SOM produisent des représentations discrètes de faible dimension des données d'entrée de haute dimension. QE est déterminée à partir des résultats du processus d'apprentissage non supervisé du SOM et des données d'entrée. SOM-QE d'une série chronologique d'images peut être utilisé comme indicateur de changements dans la série chronologique. Pour configurer SOM, on détermine la taille de la carte, la distance du voisinage, le rythme d'apprentissage et le nombre d'itérations dans le processus d'apprentissage. La combinaison de ces paramètres, qui donne la valeur la plus faible de QE, est considérée comme le jeu de paramètres optimal et est utilisée pour transformer l'ensemble de données. C'est l'utilisation de l'assouplissement quantitatif. La nouveauté de la technique SOM-QE est quadruple : d'abord dans l'usage. SOM-QE utilise un SOM pour déterminer la QE de différentes images - typiquement, dans un ensemble de données de séries temporelles - contrairement à l'utilisation traditionnelle où différents SOMs sont appliqués sur un ensemble de données. Deuxièmement, la valeur SOM-QE est introduite pour mesurer l'uniformité de l'image. Troisièmement, la valeur SOM-QE devient une étiquette spéciale et unique pour l'image dans l'ensemble de données et quatrièmement, cette étiquette est utilisée pour suivre les changements qui se produisent dans les images suivantes de la même scène. Ainsi, SOM-QE fournit une mesure des variations à l'intérieur de l'image à une instance dans le temps, et lorsqu'il est comparé aux valeurs des images subséquentes de la même scène, il révèle une visualisation transitoire des changements dans la scène à l'étude. Dans cette recherche, l'approche a été appliquée à l'imagerie artificielle, médicale et géographique pour démontrer sa performance. Les scientifiques et les ingénieurs s'intéressent aux changements qui se produisent dans les scènes géographiques d'intérêt, comme la construction de nouveaux bâtiments dans une ville ou le recul des lésions dans les images médicales. La technique SOM-QE offre un nouveau moyen de détection automatique de la croissance dans les espaces urbains ou de la progression des maladies, fournissant des informations opportunes pour une planification ou un traitement approprié. Dans ce travail, il est démontré que SOM-QE peut capturer de très petits changements dans les images. Les résultats confirment également qu'il est rapide et moins coûteux de faire la distinction entre le contenu modifié et le contenu inchangé dans les grands ensembles de données d'images. La corrélation de Pearson a confirmé qu'il y avait des corrélations statistiquement significatives entre les valeurs SOM-QE et les données réelles de vérité de terrain. Sur le plan de l'évaluation, cette technique a donné de meilleurs résultats que les autres approches existantes. Ce travail est important car il introduit une nouvelle façon d'envisager la détection rapide et automatique des changements, même lorsqu'il s'agit de petits changements locaux dans les images. Il introduit également une nouvelle méthode de détermination de QE, et les données qu'il génère peuvent être utilisées pour prédire les changements dans un ensemble de données de séries chronologiques. / A new approach for image processing, dubbed SOM-QE, that exploits the quantization error (QE) from self-organizing maps (SOM) is proposed in this thesis. SOM produce low-dimensional discrete representations of high-dimensional input data. QE is determined from the results of the unsupervised learning process of SOM and the input data. SOM-QE from a time-series of images can be used as an indicator of changes in the time series. To set-up SOM, a map size, the neighbourhood distance, the learning rate and the number of iterations in the learning process are determined. The combination of these parameters that gives the lowest value of QE, is taken to be the optimal parameter set and it is used to transform the dataset. This has been the use of QE. The novelty in SOM-QE technique is fourfold: first, in the usage. SOM-QE employs a SOM to determine QE for different images - typically, in a time series dataset - unlike the traditional usage where different SOMs are applied on one dataset. Secondly, the SOM-QE value is introduced as a measure of uniformity within the image. Thirdly, the SOM-QE value becomes a special, unique label for the image within the dataset and fourthly, this label is used to track changes that occur in subsequent images of the same scene. Thus, SOM-QE provides a measure of variations within the image at an instance in time, and when compared with the values from subsequent images of the same scene, it reveals a transient visualization of changes in the scene of study. In this research the approach was applied to artificial, medical and geographic imagery to demonstrate its performance. Changes that occur in geographic scenes of interest, such as new buildings being put up in a city or lesions receding in medical images are of interest to scientists and engineers. The SOM-QE technique provides a new way for automatic detection of growth in urban spaces or the progressions of diseases, giving timely information for appropriate planning or treatment. In this work, it is demonstrated that SOM-QE can capture very small changes in images. Results also confirm it to be fast and less computationally expensive in discriminating between changed and unchanged contents in large image datasets. Pearson's correlation confirmed that there was statistically significant correlations between SOM-QE values and the actual ground truth data. On evaluation, this technique performed better compared to other existing approaches. This work is important as it introduces a new way of looking at fast, automatic change detection even when dealing with small local changes within images. It also introduces a new method of determining QE, and the data it generates can be used to predict changes in a time series dataset.

Submitted by Cássia Santos (cassia.bcufg@gmail.com) on 2017-07-05T12:29:15Z No. of bitstreams: 2 Dissertação - Ricardo Henrique Fonseca Alves - 2017.pdf: 7774118 bytes, checksum: a675bf30443ae9eb20be6a3e5623bb4a (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-07-10T14:30:07Z (GMT) No. of bitstreams: 2 Dissertação - Ricardo Henrique Fonseca Alves - 2017.pdf: 7774118 bytes, checksum: a675bf30443ae9eb20be6a3e5623bb4a (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-07-10T14:30:07Z (GMT). No. of bitstreams: 2 Dissertação - Ricardo Henrique Fonseca Alves - 2017.pdf: 7774118 bytes, checksum: a675bf30443ae9eb20be6a3e5623bb4a (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-03-09 / Conselho Nacional de Pesquisa e Desenvolvimento Científico e Tecnológico - CNPq / The main goal of this work is to propose a methodology for the selection of 51 consumers in Nova Veneza-GO connected to two transformers in the pre-Smart Grid network. The methodology consists of ten stages ranging from the grouping of consumers with the same power consumption profile using a neural network, that is, a Non Parametric Self-Organizing Map (PSOM), until the complete and optimal allocation of financial resources through of an Integer Linear Programming. We obtained 12 different groups (clusters) of consumers of the two transformers with the same power consumption profile using the network PSOM algorithm. This grouping (clustering) was considered in the dimensioning and design of Photovoltaic Systems Connected to the Grid (Grid-Tie Systems) using three different computational tools, among them, an approach based on the PVSyst software, trial version V6.39. In addition, a study of Economic Engineering was carried out to expand the R&D pilot project aiming at the implementation of Grid Tie Systems for all the consumers of Nova Veneza-GO and Goiânia-GO, considering consumption data available by Celg-D and also considering two different scenarios based on the implementation of photovoltaic systems with and without government incentive. An Economic Engineering analysis was performed considering that 1%, 5%, 10%, 20%, 30% and 100% of Consumer Units (UC) adhere to the implantation of solar systems in Goiânia-GO. Environmental results were found for the city of Nova Veneza-GO and Goiânia-GO, evidencing an expressive reduction in CO2 emissions and a great saving of water. / O principal objetivo deste trabalho é propor uma metodologia para a escolha de 51 Unidades Consumidoras (UC) em Nova Veneza-GO ligados a dois transformadores pertencentes a uma rede pré Smart Grid localizada na cidade. A metodologia consiste de dez etapas que vão desde o agrupamento de consumidores com mesmo perfil de consumo de energia elétrica utilizando uma rede neural PSOM (do inglês: Non Parametric Self-Organizing Map), incluindo a realização de alocação de recursos financeiros de forma otimizada por meio de Programação Linear Inteira. Utilizando a rede PSOM, foi possível agrupar os consumidores dos dois transformadores em 12 grupos distintos com mesmo “perfil de consumo”. Esse agrupamento foi importante para o dimensionamento de Sistemas Fotovoltaicos Conectados à Rede Elétrica (Sistemas Grid-Tie) utilizando diferentes ferramentas computacionais, dentre elas, o software PVSyst na versão trial V6.39. Adicionalmente, foi feito um estudo de Engenharia Econômica visando a implantação de Sistemas Fotovoltaicos Conectados à Rede Elétrica para todos os consumidores de Nova Veneza-GO e de Goiânia-GO, considerando dados de consumo disponibilizados pela concessionária local e também considerando dois diferentes cenários: implantação de sistemas fotovoltaicos com e sem incentivo do governo. Foi realizada ainda uma análise de Engenharia Econômica considerando uma adesão em Goiânia-GO de 1%, 5%, 10%, 20%, 30% e 100% das Unidades Consumidoras (UC). Resultados ambientais foram encontrados para a cidade de Nova Veneza-GO e Goiânia-GO, evidenciando uma redução expressiva na emissão de CO2 e uma grande economia de água.

Data mining is about the search for relationships and global patterns in large databases that are increasing in size. Data mining is beneficial for anyone who has a huge amount of data, for example, customer and business data, transaction, marketing, financial, manufacturing and web data etc. The results of data mining are also referred to as knowledge in the form of rules, regularities and constraints. Rule mining is one of the popular data mining methods since rules provide concise statements of potentially important information that is easily understood by end users and also actionable patterns. At present rule mining has received a good deal of attention and enthusiasm from data mining researchers since rule mining is capable of solving many data mining problems such as classification, association, customer profiling, summarization, segmentation and many others. This thesis makes several contributions by proposing rule mining methods using genetic algorithms and neural networks. The thesis first proposes rule mining methods using a genetic algorithm. These methods are based on an integrated framework but capable of mining three major classes of rules. Moreover, the rule mining processes in these methods are controlled by tuning of two data mining measures such as support and confidence. The thesis shows how to build data mining predictive models using the resultant rules of the proposed methods. Another key contribution of the thesis is the proposal of rule mining methods using supervised neural networks. The thesis mathematically analyses the Widrow-Hoff learning algorithm of a single-layered neural network, which results in a foundation for rule mining algorithms using single-layered neural networks. Three rule mining algorithms using single-layered neural networks are proposed for the three major classes of rules on the basis of the proposed theorems. The thesis also looks at the problem of rule mining where user guidance is absent. The thesis proposes a guided rule mining system to overcome this problem. The thesis extends this work further by comparing the performance of the algorithm used in the proposed guided rule mining system with Apriori data mining algorithm. Finally, the thesis studies the Kohonen self-organization map as an unsupervised neural network for rule mining algorithms. Two approaches are adopted based on the way of self-organization maps applied in rule mining models. In the first approach, self-organization map is used for clustering, which provides class information to the rule mining process. In the second approach, automated rule mining takes the place of trained neurons as it grows in a hierarchical structure.