Interfaces for Intelligent Environments

Presentaciones de la asignatura Interfaces para Entornos Inteligentes del Máster en Tecnologías de la Informática/Machine Learning and Data Mining.

INVESTIGATIONS INTO MECHANISMS UNDERLYING EXTREME WAVE FORMATIONS AND COMPUTATIONALLY INTENSIVE SIMULATIONS

Various mechanisms have been proposed to explain extreme waves or rogue waves in an oceanic environment including directional focusing, dispersive focusing, wave-current interaction, and nonlinear modulational instability. The Benjamin-Feir instability (nonlinear modulational instability), however, is considered to be one of the primary mechanisms for rogue-wave occurrence. The nonlinear Schrodinger equation is a well-established approximate model based on the same assumptions as required for the derivation of the Benjamin-Feir theory. Solutions of the nonlinear Schrodinger equation, including new rogue-wave type solutions are presented in the author's dissertation work. The solutions are obtained by using a predictive eigenvalue map based predictor-corrector procedure developed by the author. Features of the predictive map are explored and the influences of certain parameter variations are investigated. The solutions are rescaled to match the length scales of waves generated in a wave tank. Based on the information provided by the map and the details of physical scaling, a framework is developed that can serve as a basis for experimental investigations into a variety of extreme waves as well localizations in wave fields. To derive further fundamental insights into the complexity of extreme wave conditions, Smoothed Particle Hydrodynamics (SPH) simulations are carried out on an advanced Graphic Processing Unit (GPU) based parallel computational platform. Free surface gravity wave simulations have successfully characterized water-wave dispersion in the SPH model while demonstrating extreme energy focusing and wave growth in both linear and nonlinear regimes. A virtual wave tank is simulated wherein wave motions can be excited from either side. Focusing of several wave trains and isolated waves has been simulated. With properly chosen parameters, dispersion effects are observed causing a chirped wave train to focus and exhibit growth. By using the insights derived from the study of the nonlinear Schrodinger equation, modulational instability or self-focusing has been induced in a numerical wave tank and studied through several numerical simulations. Due to the inherent dissipative nature of SPH models, simulating persistent progressive waves can be problematic. This issue has been addressed and an observation-based solution has been provided. The efficacy of SPH in modeling wave focusing can be critical to further our understanding and predicting extreme wave phenomena through simulations. A deeper understanding of the mechanisms underlying extreme energy localization phenomena can help facilitate energy harnessing and serve as a basis to predict and mitigate the impact of energy focusing.

MINING AND VERIFICATION OF TEMPORAL EVENTS WITH APPLICATIONS IN COMPUTER MICRO-ARCHITECTURE RESEARCH

Computer simulation programs are essential tools for scientists and engineers to understand a particular system of interest. As expected, the complexity of the software increases with the depth of the model used. In addition to the exigent demands of software engineering, verification of simulation programs is especially challenging because the models represented are complex and ridden with unknowns that will be discovered by developers in an iterative process. To manage such complexity, advanced verification techniques for continually matching the intended model to the implemented model are necessary. Therefore, the main goal of this research work is to design a useful verification and validation framework that is able to identify model representation errors and is applicable to generic simulators. The framework that was developed and implemented consists of two parts. The first part is First-Order Logic Constraint Specification Language (FOLCSL) that enables users to specify the invariants of a model under consideration. From the first-order logic specification, the FOLCSL translator automatically synthesizes a verification program that reads the event trace generated by a simulator and signals whether all invariants are respected. The second part consists of mining the temporal flow of events using a newly developed representation called State Flow Temporal Analysis Graph (SFTAG). While the first part seeks an assurance of implementation correctness by checking that the model invariants hold, the second part derives an extended model of the implementation and hence enables a deeper understanding of what was implemented. The main application studied in this work is the validation of the timing behavior of micro-architecture simulators. The study includes SFTAGs generated for a wide set of benchmark programs and their analysis using several artificial intelligence algorithms. This work improves the computer architecture research and verification processes as shown by the case studies and experiments that have been conducted.

Ensemble Stream Model for Data-Cleaning in Sensor Networks

Ensemble Stream Modeling and Data-cleaning are sensor information processing systems have different training and testing methods by which their goals are cross-validated. This research examines a mechanism, which seeks to extract novel patterns by generating ensembles from data. The main goal of label-less stream processing is to process the sensed events to eliminate the noises that are uncorrelated, and choose the most likely model without over fitting thus obtaining higher model confidence. Higher quality streams can be realized by combining many short streams into an ensemble which has the desired quality. The framework for the investigation is an existing data mining tool. First, to accommodate feature extraction such as a bush or natural forest-fire event we make an assumption of the burnt area (BA*), sensed ground truth as our target variable obtained from logs. Even though this is an obvious model choice the results are disappointing. The reasons for this are two: One, the histogram of fire activity is highly skewed. Two, the measured sensor parameters are highly correlated. Since using non descriptive features does not yield good results, we resort to temporal features. By doing so we carefully eliminate the averaging effects; the resulting histogram is more satisfactory and conceptual knowledge is learned from sensor streams. Second is the process of feature induction by cross-validating attributes with single or multi-target variables to minimize training error. We use F-measure score, which combines precision and accuracy to determine the false alarm rate of fire events. The multi-target data-cleaning trees use information purity of the target leaf-nodes to learn higher order features. A sensitive variance measure such as f-test is performed during each node’s split to select the best attribute. Ensemble stream model approach proved to improve when using complicated features with a simpler tree classifier. The ensemble framework for data-cleaning and the enhancements to quantify quality of fitness (30% spatial, 10% temporal, and 90% mobility reduction) of sensor led to the formation of streams for sensor-enabled applications. Which further motivates the novelty of stream quality labeling and its importance in solving vast amounts of real-time mobile streams generated today.

Using multiple GPUs to accelerate string searching for digital forensic analysis

String searching within a large corpus of data is an important component of digital forensic (DF) analysis techniques such as file carving. The continuing increase in capacity of consumer storage devices requires corresponding im-provements to the performance of string searching techniques. As string search-ing is a trivially-parallelisable problem, GPGPU approaches are a natural fit – but previous studies have found that local storage presents an insurmountable performance bottleneck. We show that this need not be the case with modern hardware, and demonstrate substantial performance improvements from the use of single and multiple GPUs when searching for strings within a typical forensic disk image.

Caracterización sociodemográfica de las enfermedades huérfanas en Colombia

Las enfermedades huérfanas en Colombia, se definen como aquellas crónicamente debilitantes, que amenazan la vida, de baja prevalencia (menor 1/5000) y alta complejidad. Se estima que a nivel mundial existen entre 6000 a 8000 enfermedades raras diferentes(1). Varios países a nivel mundial individual o colectivamente, en los últimos años han creado políticas e incentivos para la investigación y protección de los pacientes con enfermedades raras. Sin embargo, a pesar del creciente número de publicaciones; la información sobre su etiología, fisiología, historia natural y datos epidemiológicos persiste escasa o ausente. Los registros de pacientes, son una valiosa herramienta para la caracterización de las enfermedades, su manejo y desenlaces con o sin tratamiento. Permiten mejorar políticas de salud pública y cuidado del paciente, contribuyendo a mejorar desenlaces sociales, económicos y de calidad de vida. En Colombia, bajo el decreto 1954 de 2012 y las resoluciones 3681 de 2013 y 0430 de 2013 se creó el fundamento legal para la creación de un registro nacional de enfermedades huérfanas. El presente estudio busca determinar la caracterización socio-demográfica y la prevalencia de las enfermedades huérfanas en Colombia en el periodo 2013. Métodos: Se realizó un estudio observacional de corte transversal de fuente secundaria sobre pacientes con enfermedades huérfanas en el territorio nacional; basándose en el registro nacional de enfermedades huérfanas obtenido por el Ministerio de Salud y Protección Social en el periodo 2013 bajo la normativa del decreto 1954 de 2012 y las resoluciones 3681 de 2013 y 0430 de 2013. Las bases de datos obtenidas fueron re-categorizadas en Excel versión 15.17 para la extracción de datos y su análisis estadístico posterior, fue realizado en el paquete estadístico para las ciencias sociales (SPSS v.20, Chicago, IL). Resultados: Se encontraron un total de 13173 pacientes con enfermedades huérfanas para el 2013. De estos, el 53.96% (7132) eran de género femenino y el 46.03% (6083) masculino; la mediana de la edad fue de 28 años con un rango inter-cuartil de 39 años, el 9% de los pacientes presentaron discapacidad. El registro contenía un total de 653 enfermedades huérfanas; el 34% del total de las enfermedades listadas en nuestro país (2). Las patologías más frecuentes fueron el Déficit Congénito del Factor VIII, Miastenia Grave, Enfermedad de Von Willebrand, Estatura Baja por Anomalía de Hormona de Crecimiento y Displasia Broncopulmonar. Discusión: Se estimó que aproximadamente 3.3 millones de colombianos debían tener una enfermedad huérfana para el 2013. El registro nacional logró recolectar datos de 13173 (0.4%). Este bajo número de pacientes, marca un importante sub-registro que se debe al uso de los códigos CIE-10, desconocimiento del personal de salud frente a las enfermedades huérfanas y clasificación errónea de los pacientes. Se encontraron un total de 653 enfermedades, un 34% de las enfermedades reportadas en el listado nacional de enfermedades huérfanas (2) y un 7% del total de enfermedades reportadas en ORPHANET para el periodo 2013 (3). Conclusiones: La recolección de datos y la sensibilización sobre las enfermedades huérfanas al personal de salud, es una estrategia de vital importancia para el diagnóstico temprano, medidas específicas de control e intervenciones de los pacientes. El identificar apropiadamente a los pacientes con este tipo de patologías, permite su ingreso en el registro y por ende mejora el sub-registro de datos. Sin embargo, cabe aclarar que el panorama ideal sería, el uso de un sistema de recolección diferente al CIE-10 y que abarque en mayor medida la totalidad de las enfermedades huérfanas.

Machine Learning Unsupervised Methods in the Design of an On-board Health Monitoring System for Satellite Applications

The dissertation starts by providing a description of the phenomena related to the increasing importance recently acquired by satellite applications. The spread of such technology comes with implications, such as an increase in maintenance cost, from which derives the interest in developing advanced techniques that favor an augmented autonomy of spacecrafts in health monitoring. Machine learning techniques are widely employed to lay a foundation for effective systems specialized in fault detection by examining telemetry data. Telemetry consists of a considerable amount of information; therefore, the adopted algorithms must be able to handle multivariate data while facing the limitations imposed by on-board hardware features. In the framework of outlier detection, the dissertation addresses the topic of unsupervised machine learning methods. In the unsupervised scenario, lack of prior knowledge of the data behavior is assumed. In the specific, two models are brought to attention, namely Local Outlier Factor and One-Class Support Vector Machines. Their performances are compared in terms of both the achieved prediction accuracy and the equivalent computational cost. Both models are trained and tested upon the same sets of time series data in a variety of settings, finalized at gaining insights on the effect of the increase in dimensionality. The obtained results allow to claim that both models, combined with a proper tuning of their characteristic parameters, successfully comply with the role of outlier detectors in multivariate time series data. Nevertheless, under this specific context, Local Outlier Factor results to be outperforming One-Class SVM, in that it proves to be more stable over a wider range of input parameter values. This property is especially valuable in unsupervised learning since it suggests that the model is keen to adapting to unforeseen patterns.

Emprego de regras de associação para extração de padrões mercadológicos de touros Nelore com avaliação genética

Objetivou-se com este trabalho utilizar regras de associação para identificar forças de mercado que regem a comercialização de touros com avaliação genética pelo programa Nelore Brasil. Essas regras permitem evidenciar padrões implícitos nas transações de grandes bases de dados, indicando causas e efeitos determinantes da oferta e comercialização de touros. Na análise foram considerados 19.736 registros de touros comercializados, 17 fazendas e 15 atributos referentes às diferenças esperadas nas progênies dos reprodutores, local e época da venda. Utilizou-se um sistema com interface gráfica usuário-dirigido que permite geração e seleção interativa de regras de associação. Análise de Pareto foi aplicada para as três medidas objetivas (suporte, confiança e lift) que acompanham cada uma das regras de associação, para validação das mesmas. Foram geradas 2.667 regras de associação, 164 consideradas úteis pelo usuário e 107 válidas para lift ≥ 1,0505. As fazendas participantes do programa Nelore Brasil apresentam especializações na oferta de touros, segundo características para habilidade materna, ganho de peso, fertilidade, precocidade sexual, longevidade, rendimento e terminação de carcaça. Os perfis genéticos dos touros são diferentes para as variedades padrão e mocho. Algumas regiões brasileiras são nichos de mercado para touros sem registro genealógico. A análise de evolução de mercado sugere que o mérito genético total, índice oficial do programa Nelore Brasil, tornou-se um importante índice para comercialização dos touros. Com o uso das regras de associação, foi possível descobrir forças do mercado e identificar combinações de atributos genéticos, geográficos e temporais que determinam a comercialização de touros no programa Nelore Brasil.

Support vector machines for detecting age-related changes in running kinematics

Age-related changes in running kinematics have been reported in the literature using classical inferential statistics. However, this approach has been hampered by the increased number of biomechanical gait variables reported and subsequently the lack of differences presented in these studies. Data mining techniques have been applied in recent biomedical studies to solve this problem using a more general approach. In the present work, we re-analyzed lower extremity running kinematic data of 17 young and 17 elderly male runners using the Support Vector Machine (SVM) classification approach. In total, 31 kinematic variables were extracted to train the classification algorithm and test the generalized performance. The results revealed different accuracy rates across three different kernel methods adopted in the classifier, with the linear kernel performing the best. A subsequent forward feature selection algorithm demonstrated that with only six features, the linear kernel SVM achieved 100% classification performance rate, showing that these features provided powerful combined information to distinguish age groups. The results of the present work demonstrate potential in applying this approach to improve knowledge about the age-related differences in running gait biomechanics and encourages the use of the SVM in other clinical contexts. (C) 2010 Elsevier Ltd. All rights reserved.

Efficiency and feasibility of product disassembly: A case-based study

The productivity associated with commonly available disassembly methods today seldomly makes disassembly the preferred end-of-life solution for massive take back product streams. Systematic reuse of parts or components, or recycling of pure material fractions are often not achievable in an economically sustainable way. In this paper a case-based review of current disassembly practices is used to analyse the factors influencing disassembly feasibility. Data mining techniques were used to identify major factors influencing the profitability of disassembly operations. Case characteristics such as involvement of the product manufacturer in the end-of-life treatment and continuous ownership are some of the important dimensions. Economic models demonstrate that the efficiency of disassembly operations should be increased an order of magnitude to assure the competitiveness of ecologically preferred, disassembly oriented end-of-life scenarios for large waste of electric and electronic equipment (WEEE) streams. Technological means available to increase the productivity of the disassembly operations are summarized. Automated disassembly techniques can contribute to the robustness of the process, but do not allow to overcome the efficiency gap if not combined with appropriate product design measures. Innovative, reversible joints, collectively activated by external trigger signals, form a promising approach to low cost, mass disassembly in this context. A short overview of the state-of-the-art in the development of such self-disassembling joints is included. (c) 2008 CIRP.

In silico analysis of candidate genes involved in light sensing and signal transduction pathways in soybean

Several aspects of photoperception and light signal transduction have been elucidated by studies with model plants. However, the information available for economically important crops, such as Fabaceae species, is scarce. In order to incorporate the existing genomic tools into a strategy to advance soybean research, we have investigated publicly available expressed sequence tag ( EST) sequence databases in order to identify Glycine max sequences related to genes involved in light-regulated developmental control in model plants. Approximately 38,000 sequences from open-access databases were investigated, and all bona fide and putative photoreceptor gene families were found in soybean sequence databases. We have identified G. max orthologs for several families of transcriptional regulators and cytoplasmic proteins mediating photoreceptor-induced responses, although some important Arabidopsis phytochrome-signaling components are absent. Moreover, soybean and Arabidopsis gene-family homologs appear to have undergone a distinct expansion process in some cases. We propose a working model of light perception, signal transduction and response-eliciting in G. max, based on the identified key components from Arabidopsis. These results demonstrate the power of comparative genomics between model systems and crop species to elucidate several aspects of plant physiology and metabolism.

A strategy to fill gaps in soil survey over large spatial extents: an example from the Murray-Darling basin of Australia

We re-mapped the soils of the Murray-Darling Basin (MDB) in 1995-1998 with a minimum of new fieldwork, making the most out of existing data. We collated existing digital soil maps and used inductive spatial modelling to predict soil types from those maps combined with environmental predictor variables. Lithology, Landsat Multi Spectral Scanner (Landsat MSS), the 9-s digital elevation model (DEM) of Australia and derived terrain attributes, all gridded to 250-m pixels, were the predictor variables. Because the basin-wide datasets were very large data mining software was used for modelling. Rule induction by data mining was also used to define the spatial domain of extrapolation for the extension of soil-landscape models from existing soil maps. Procedures to estimate the uncertainty associated with the predictions and quality of information for the new soil-landforms map of the MDB are described. (C) 2002 Elsevier Science B.V. All rights reserved.

Australia-wide predictions of soil properties using decision trees

This paper describes the construction of Australia-wide soil property predictions from a compiled national soils point database. Those properties considered include pH, organic carbon, total phosphorus, total nitrogen, thickness. texture, and clay content. Many of these soil properties are used directly in environmental process modelling including global climate change models. Models are constructed at the 250-m resolution using decision trees. These relate the soil property to the environment through a suite of environmental predictors at the locations where measurements are observed. These models are then used to extend predictions to the continental extent by applying the rules derived to the exhaustively available environmental predictors. The methodology and performance is described in detail for pH and summarized for other properties. Environmental variables are found to be important predictors, even at the 250-m resolution at which they are available here as they can describe the broad changes in soil property.

DNA polymorphisms as tools for spinal cord injury research

Study Design: Data mining of single nucleotide polymorphisms (SNPs) in gene pathways related to spinal cord injury (SCI). Objectives: To identify gene polymorphisms putatively implicated with neuronal damage evolution pathways, potentially useful to SCI study. Setting: Departments of Psychiatry and Orthopedics, Faculdade de Medicina, Universidade de Sao Paulo, Brazil. Methods: Genes involved with processes related to SCI, such as apoptosis, inflammatory response, axonogenesis, peripheral nervous system development and axon ensheathment, were determined by evaluating the `Biological Process` annotation of Gene Ontology (GO). Each gene of these pathways was mapped using MapViewer, and gene coordinates were used to identify their polymorphisms in the SNP database. As a proof of concept, the frequency of subset of SNPs, located in four genes (ALOX12, APOE, BDNF and NINJ1) was evaluated in the DNA of a group of 28 SCI patients and 38 individuals with no SC lesions. Results: We could identify a total of 95 276 SNPs in a set of 588 genes associated with the selected GO terms, including 3912 nucleotide alterations located in coding regions of genes. The five non-synonymous SNPs genotyped in our small group of patients, showed a significant frequency, reinforcing their potential use for the investigation of SCI evolution. Conclusion: Despite the importance of SNPs in many aspects of gene expression and protein activity, these gene alterations have not been explored in SCI research. Here we describe a set of potentially useful SNPs, some of which could underlie the genetic mechanisms involved in the post trauma spinal cord damage.

The application of adaptive partitioned random search in feature selection problem

Feature selection is one of important and frequently used techniques in data preprocessing. It can improve the efficiency and the effectiveness of data mining by reducing the dimensions of feature space and removing the irrelevant and redundant information. Feature selection can be viewed as a global optimization problem of finding a minimum set of M relevant features that describes the dataset as well as the original N attributes. In this paper, we apply the adaptive partitioned random search strategy into our feature selection algorithm. Under this search strategy, the partition structure and evaluation function is proposed for feature selection problem. This algorithm ensures the global optimal solution in theory and avoids complete randomness in search direction. The good property of our algorithm is shown through the theoretical analysis.