Autonomous classification models in ubiquitous environments

Stream-mining approach is defined as a set of cutting-edge techniques designed to process streams of data in real time, in order to extract knowledge. In the particular case of classification, stream-mining has to adapt its behaviour to the volatile underlying data distributions, what has been called concept drift. Moreover, it is important to note that concept drift may lead to situations where predictive models become invalid and have therefore to be updated to represent the actual concepts that data poses. In this context, there is a specific type of concept drift, known as recurrent concept drift, where the concepts represented by data have already appeared in the past. In those cases the learning process could be saved or at least minimized by applying a previously trained model. This could be extremely useful in ubiquitous environments that are characterized by the existence of resource constrained devices. To deal with the aforementioned scenario, meta-models can be used in the process of enhancing the drift detection mechanisms used by data stream algorithms, by representing and predicting when the change will occur. There are some real-world situations where a concept reappears, as in the case of intrusion detection systems (IDS), where the same incidents or an adaptation of them usually reappear over time. In these environments the early prediction of drift by means of a better knowledge of past models can help to anticipate to the change, thus improving efficiency of the model regarding the training instances needed. By means of using meta-models as a recurrent drift detection mechanism, the ability to share concepts representations among different data mining processes is open. That kind of exchanges could improve the accuracy of the resultant local model as such model may benefit from patterns similar to the local concept that were observed in other scenarios, but not yet locally. This would also improve the efficiency of training instances used during the classification process, as long as the exchange of models would aid in the application of already trained recurrent models, that have been previously seen by any of the collaborative devices. Which it is to say that the scope of recurrence detection and representation is broaden. In fact the detection, representation and exchange of concept drift patterns would be extremely useful for the law enforcement activities fighting against cyber crime. Being the information exchange one of the main pillars of cooperation, national units would benefit from the experience and knowledge gained by third parties. Moreover, in the specific scope of critical infrastructures protection it is crucial to count with information exchange mechanisms, both from a strategical and technical scope. The exchange of concept drift detection schemes in cyber security environments would aid in the process of preventing, detecting and effectively responding to threads in cyber space. Furthermore, as a complement of meta-models, a mechanism to assess the similarity between classification models is also needed when dealing with recurrent concepts. In this context, when reusing a previously trained model a rough comparison between concepts is usually made, applying boolean logic. The introduction of fuzzy logic comparisons between models could lead to a better efficient reuse of previously seen concepts, by applying not just equal models, but also similar ones. This work faces the aforementioned open issues by means of: the MMPRec system, that integrates a meta-model mechanism and a fuzzy similarity function; a collaborative environment to share meta-models between different devices; a recurrent drift generator that allows to test the usefulness of recurrent drift systems, as it is the case of MMPRec. Moreover, this thesis presents an experimental validation of the proposed contributions using synthetic and real datasets.

Models for Internal Quality Fruit Sorting, Based on Time- Domain Laser Reflectance Spectroscopy (TDRS)

Time domain laser reflectance spectroscopy (TDRS) was applied for the first time to evaluate internal fruit quality. This technique, known in medicine-related knowledge areas, has not been used before in agricultural or food research. It allows the simultaneous non-destructive measuring of two optical characteristics of the tissues: light scattering and absorption. Models to measure firmness, sugar & acid contents in kiwifruit, tomato, apple, peach, nectarine and other fruits were built using sequential statistical techniques: principal component analysis, multiple stepwise linear regression, clustering and discriminant analysis. Consistent correlations were established between the two parameters measured with TDRS, i.e. absorption & transport scattering coefficients, with chemical constituents (sugars and acids) and firmness, respectively. Classification models were built to sort fruits into three quality grades, according to their firmness, soluble solids and acidity.

Multi-dimensional classification using Bayesian networks for stationary and evolving streaming data

Hoy en día, con la evolución continua y rápida de las tecnologías de la información y los dispositivos de computación, se recogen y almacenan continuamente grandes volúmenes de datos en distintos dominios y a través de diversas aplicaciones del mundo real. La extracción de conocimiento útil de una cantidad tan enorme de datos no se puede realizar habitualmente de forma manual, y requiere el uso de técnicas adecuadas de aprendizaje automático y de minería de datos. La clasificación es una de las técnicas más importantes que ha sido aplicada con éxito a varias áreas. En general, la clasificación se compone de dos pasos principales: en primer lugar, aprender un modelo de clasificación o clasificador a partir de un conjunto de datos de entrenamiento, y en segundo lugar, clasificar las nuevas instancias de datos utilizando el clasificador aprendido. La clasificación es supervisada cuando todas las etiquetas están presentes en los datos de entrenamiento (es decir, datos completamente etiquetados), semi-supervisada cuando sólo algunas etiquetas son conocidas (es decir, datos parcialmente etiquetados), y no supervisada cuando todas las etiquetas están ausentes en los datos de entrenamiento (es decir, datos no etiquetados). Además, aparte de esta taxonomía, el problema de clasificación se puede categorizar en unidimensional o multidimensional en función del número de variables clase, una o más, respectivamente; o también puede ser categorizado en estacionario o cambiante con el tiempo en función de las características de los datos y de la tasa de cambio subyacente. A lo largo de esta tesis, tratamos el problema de clasificación desde tres perspectivas diferentes, a saber, clasificación supervisada multidimensional estacionaria, clasificación semisupervisada unidimensional cambiante con el tiempo, y clasificación supervisada multidimensional cambiante con el tiempo. Para llevar a cabo esta tarea, hemos usado básicamente los clasificadores Bayesianos como modelos. La primera contribución, dirigiéndose al problema de clasificación supervisada multidimensional estacionaria, se compone de dos nuevos métodos de aprendizaje de clasificadores Bayesianos multidimensionales a partir de datos estacionarios. Los métodos se proponen desde dos puntos de vista diferentes. El primer método, denominado CB-MBC, se basa en una estrategia de envoltura de selección de variables que es voraz y hacia delante, mientras que el segundo, denominado MB-MBC, es una estrategia de filtrado de variables con una aproximación basada en restricciones y en el manto de Markov. Ambos métodos han sido aplicados a dos problemas reales importantes, a saber, la predicción de los inhibidores de la transcriptasa inversa y de la proteasa para el problema de infección por el virus de la inmunodeficiencia humana tipo 1 (HIV-1), y la predicción del European Quality of Life-5 Dimensions (EQ-5D) a partir de los cuestionarios de la enfermedad de Parkinson con 39 ítems (PDQ-39). El estudio experimental incluye comparaciones de CB-MBC y MB-MBC con los métodos del estado del arte de la clasificación multidimensional, así como con métodos comúnmente utilizados para resolver el problema de predicción de la enfermedad de Parkinson, a saber, la regresión logística multinomial, mínimos cuadrados ordinarios, y mínimas desviaciones absolutas censuradas. En ambas aplicaciones, los resultados han sido prometedores con respecto a la precisión de la clasificación, así como en relación al análisis de las estructuras gráficas que identifican interacciones conocidas y novedosas entre las variables. La segunda contribución, referida al problema de clasificación semi-supervisada unidimensional cambiante con el tiempo, consiste en un método nuevo (CPL-DS) para clasificar flujos de datos parcialmente etiquetados. Los flujos de datos difieren de los conjuntos de datos estacionarios en su proceso de generación muy rápido y en su aspecto de cambio de concepto. Es decir, los conceptos aprendidos y/o la distribución subyacente están probablemente cambiando y evolucionando en el tiempo, lo que hace que el modelo de clasificación actual sea obsoleto y deba ser actualizado. CPL-DS utiliza la divergencia de Kullback-Leibler y el método de bootstrapping para cuantificar y detectar tres tipos posibles de cambio: en las predictoras, en la a posteriori de la clase o en ambas. Después, si se detecta cualquier cambio, un nuevo modelo de clasificación se aprende usando el algoritmo EM; si no, el modelo de clasificación actual se mantiene sin modificaciones. CPL-DS es general, ya que puede ser aplicado a varios modelos de clasificación. Usando dos modelos diferentes, el clasificador naive Bayes y la regresión logística, CPL-DS se ha probado con flujos de datos sintéticos y también se ha aplicado al problema real de la detección de código malware, en el cual los nuevos ficheros recibidos deben ser continuamente clasificados en malware o goodware. Los resultados experimentales muestran que nuestro método es efectivo para la detección de diferentes tipos de cambio a partir de los flujos de datos parcialmente etiquetados y también tiene una buena precisión de la clasificación. Finalmente, la tercera contribución, sobre el problema de clasificación supervisada multidimensional cambiante con el tiempo, consiste en dos métodos adaptativos, a saber, Locally Adpative-MB-MBC (LA-MB-MBC) y Globally Adpative-MB-MBC (GA-MB-MBC). Ambos métodos monitorizan el cambio de concepto a lo largo del tiempo utilizando la log-verosimilitud media como métrica y el test de Page-Hinkley. Luego, si se detecta un cambio de concepto, LA-MB-MBC adapta el actual clasificador Bayesiano multidimensional localmente alrededor de cada nodo cambiado, mientras que GA-MB-MBC aprende un nuevo clasificador Bayesiano multidimensional. El estudio experimental realizado usando flujos de datos sintéticos multidimensionales indica los méritos de los métodos adaptativos propuestos. ABSTRACT Nowadays, with the ongoing and rapid evolution of information technology and computing devices, large volumes of data are continuously collected and stored in different domains and through various real-world applications. Extracting useful knowledge from such a huge amount of data usually cannot be performed manually, and requires the use of adequate machine learning and data mining techniques. Classification is one of the most important techniques that has been successfully applied to several areas. Roughly speaking, classification consists of two main steps: first, learn a classification model or classifier from an available training data, and secondly, classify the new incoming unseen data instances using the learned classifier. Classification is supervised when the whole class values are present in the training data (i.e., fully labeled data), semi-supervised when only some class values are known (i.e., partially labeled data), and unsupervised when the whole class values are missing in the training data (i.e., unlabeled data). In addition, besides this taxonomy, the classification problem can be categorized into uni-dimensional or multi-dimensional depending on the number of class variables, one or more, respectively; or can be also categorized into stationary or streaming depending on the characteristics of the data and the rate of change underlying it. Through this thesis, we deal with the classification problem under three different settings, namely, supervised multi-dimensional stationary classification, semi-supervised unidimensional streaming classification, and supervised multi-dimensional streaming classification. To accomplish this task, we basically used Bayesian network classifiers as models. The first contribution, addressing the supervised multi-dimensional stationary classification problem, consists of two new methods for learning multi-dimensional Bayesian network classifiers from stationary data. They are proposed from two different points of view. The first method, named CB-MBC, is based on a wrapper greedy forward selection approach, while the second one, named MB-MBC, is a filter constraint-based approach based on Markov blankets. Both methods are applied to two important real-world problems, namely, the prediction of the human immunodeficiency virus type 1 (HIV-1) reverse transcriptase and protease inhibitors, and the prediction of the European Quality of Life-5 Dimensions (EQ-5D) from 39-item Parkinson’s Disease Questionnaire (PDQ-39). The experimental study includes comparisons of CB-MBC and MB-MBC against state-of-the-art multi-dimensional classification methods, as well as against commonly used methods for solving the Parkinson’s disease prediction problem, namely, multinomial logistic regression, ordinary least squares, and censored least absolute deviations. For both considered case studies, results are promising in terms of classification accuracy as well as regarding the analysis of the learned MBC graphical structures identifying known and novel interactions among variables. The second contribution, addressing the semi-supervised uni-dimensional streaming classification problem, consists of a novel method (CPL-DS) for classifying partially labeled data streams. Data streams differ from the stationary data sets by their highly rapid generation process and their concept-drifting aspect. That is, the learned concepts and/or the underlying distribution are likely changing and evolving over time, which makes the current classification model out-of-date requiring to be updated. CPL-DS uses the Kullback-Leibler divergence and bootstrapping method to quantify and detect three possible kinds of drift: feature, conditional or dual. Then, if any occurs, a new classification model is learned using the expectation-maximization algorithm; otherwise, the current classification model is kept unchanged. CPL-DS is general as it can be applied to several classification models. Using two different models, namely, naive Bayes classifier and logistic regression, CPL-DS is tested with synthetic data streams and applied to the real-world problem of malware detection, where the new received files should be continuously classified into malware or goodware. Experimental results show that our approach is effective for detecting different kinds of drift from partially labeled data streams, as well as having a good classification performance. Finally, the third contribution, addressing the supervised multi-dimensional streaming classification problem, consists of two adaptive methods, namely, Locally Adaptive-MB-MBC (LA-MB-MBC) and Globally Adaptive-MB-MBC (GA-MB-MBC). Both methods monitor the concept drift over time using the average log-likelihood score and the Page-Hinkley test. Then, if a drift is detected, LA-MB-MBC adapts the current multi-dimensional Bayesian network classifier locally around each changed node, whereas GA-MB-MBC learns a new multi-dimensional Bayesian network classifier from scratch. Experimental study carried out using synthetic multi-dimensional data streams shows the merits of both proposed adaptive methods.

Evaluación de la contribución y el impacto de las tecnologías del habla en la detección automática del Síndrome de la Apnea Obstructiva del Sueño = Contributions and impact assessment of speech technologies on the automatic detection of severe Obstructive Sleep Apnea syndrome

La presente Tesis analiza las posibilidades que ofrecen en la actualidad las tecnologías del habla para la detección de patologías clínicas asociadas a la vía aérea superior. El estudio del habla que tradicionalmente cubre tanto la producción como el proceso de transformación del mensaje y las señales involucradas, desde el emisor hasta alcanzar al receptor, ofrece una vía de estudio alternativa para estas patologías. El hecho de que la señal emitida no solo contiene este mensaje, sino también información acerca del locutor, ha motivado el desarrollo de sistemas orientados a la identificación y verificación de la identidad de los locutores. Estos trabajos han recibido recientemente un nuevo impulso, orientándose tanto hacia la caracterización de rasgos que son comunes a varios locutores, como a las diferencias existentes entre grabaciones de un mismo locutor. Los primeros resultan especialmente relevantes para esta Tesis dado que estos rasgos podrían evidenciar la presencia de características relacionadas con una cierta condición común a varios locutores, independiente de su identidad. Tal es el caso que se enfrenta en esta Tesis, donde los rasgos identificados se relacionarían con una de la patología particular y directamente vinculada con el sistema de físico de conformación del habla. El caso del Síndrome de Apneas Hipopneas durante el Sueno (SAHS) resulta paradigmático. Se trata de una patología con una elevada prevalencia mundo, que aumenta con la edad. Los pacientes de esta patología experimentan episodios de cese involuntario de la respiración durante el sueño, que se prolongan durante varios segundos y que se reproducen a lo largo de la noche impidiendo el correcto descanso. En el caso de la apnea obstructiva, estos episodios se deben a la imposibilidad de mantener un camino abierto a través de la vía aérea, de forma que el flujo de aire se ve interrumpido. En la actualidad, el diagnostico de estos pacientes se realiza a través de un estudio polisomnográfico, que se centra en el análisis de los episodios de apnea durante el sueño, requiriendo que el paciente permanezca en el hospital durante una noche. La complejidad y el elevado coste de estos procedimientos, unidos a las crecientes listas de espera, han evidenciado la necesidad de contar con técnicas rápidas de detección, que si bien podrían no obtener tasas tan elevadas, permitirían reorganizar las listas de espera en función del grado de severidad de la patología en cada paciente. Entre otros, los sistemas de diagnostico por imagen, así como la caracterización antropométrica de los pacientes, han evidenciado la existencia de patrones anatómicos que tendrían influencia directa sobre el habla. Los trabajos dedicados al estudio del SAHS en lo relativo a como esta afecta al habla han sido escasos y algunos de ellos incluso contradictorios. Sin embargo, desde finales de la década de 1980 se conoce la existencia de patrones específicos relativos a la articulación, la fonación y la resonancia. Sin embargo, su descripción resultaba difícilmente aprovechable a través de un sistema de reconocimiento automático, pero apuntaba la existencia de un nexo entre voz y SAHS. En los últimos anos las técnicas de procesado automático han permitido el desarrollo de sistemas automáticos que ya son capaces de identificar diferencias significativas en el habla de los pacientes del SAHS, y que los distinguen de los locutores sanos. Por contra, poco se conoce acerca de la conexión entre estos nuevos resultados, los sé que habían obtenido en el pasado y la patogénesis del SAHS. Esta Tesis continua la labor desarrollada en este ámbito considerando específicamente: el estudio de la forma en que el SAHS afecta el habla de los pacientes, la mejora en las tasas de clasificación automática y la combinación de la información obtenida con los predictores utilizados por los especialistas clínicos en sus evaluaciones preliminares. Las dos primeras tareas plantean problemas simbióticos, pero diferentes. Mientras el estudio de la conexión entre el SAHS y el habla requiere de modelos acotados que puedan ser interpretados con facilidad, los sistemas de reconocimiento se sirven de un elevado número de dimensiones para la caracterización y posterior identificación de patrones. Así, la primera tarea debe permitirnos avanzar en la segunda, al igual que la incorporación de los predictores utilizados por los especialistas clínicos. La Tesis aborda el estudio tanto del habla continua como del habla sostenida, con el fin de aprovechar las sinergias y diferencias existentes entre ambas. En el análisis del habla continua se tomo como punto de partida un esquema que ya fue evaluado con anterioridad, y sobre el cual se ha tratado la evaluación y optimización de la representación del habla, así como la caracterización de los patrones específicos asociados al SAHS. Ello ha evidenciado la conexión entre el SAHS y los elementos fundamentales de la señal de voz: los formantes. Los resultados obtenidos demuestran que el éxito de estos sistemas se debe, fundamentalmente, a la capacidad de estas representaciones para describir dichas componentes, obviando las dimensiones ruidosas o con poca capacidad discriminativa. El esquema resultante ofrece una tasa de error por debajo del 18%, sirviéndose de clasificadores notablemente menos complejos que los descritos en el estado del arte y de una única grabación de voz de corta duración. En relación a la conexión entre el SAHS y los patrones observados, fue necesario considerar las diferencias inter- e intra-grupo, centrándonos en la articulación característica del locutor, sustituyendo los complejos modelos de clasificación por el estudio de los promedios espectrales. El resultado apunta con claridad hacia ciertas regiones del eje de frecuencias, sugiriendo la existencia de un estrechamiento sistemático en la sección del tracto en la región de la orofaringe, ya prevista en la patogénesis de este síndrome. En cuanto al habla sostenida, se han reproducido los estudios realizados sobre el habla continua en grabaciones de la vocal /a/ sostenida. Los resultados son cualitativamente análogos a los anteriores, si bien en este caso las tasas de clasificación resultan ser más bajas. Con el objetivo de identificar el sentido de este resultado se reprodujo el estudio de los promedios espectrales y de la variabilidad inter e intra-grupo. Ambos estudios mostraron importantes diferencias con los anteriores que podrían explicar estos resultados. Sin embargo, el habla sostenida ofrece otras oportunidades al establecer un entorno controlado para el estudio de la fonación, que también había sido identificada como una fuente de información para la detección del SAHS. De su estudio se pudo observar que, en el conjunto de datos disponibles, no existen variaciones que pudieran asociarse fácilmente con la fonación. Únicamente aquellas dimensiones que describen la distribución de energía a lo largo del eje de frecuencia evidenciaron diferencias significativas, apuntando, una vez más, en la dirección de las resonancias espectrales. Analizados los resultados anteriores, la Tesis afronta la fusión de ambas fuentes de información en un único sistema de clasificación. Con ello es posible mejorar las tasas de clasificación, bajo la hipótesis de que la información presente en el habla continua y el habla sostenida es fundamentalmente distinta. Esta tarea se realizo a través de un sencillo esquema de fusión que obtuvo un 88.6% de aciertos en clasificación (tasa de error del 11.4%), lo que representa una mejora significativa respecto al estado del arte. Finalmente, la combinación de este clasificador con los predictores utilizados por los especialistas clínicos ofreció una tasa del 91.3% (tasa de error de 8.7%), que se encuentra dentro del margen ofrecido por esquemas más costosos e intrusivos, y que a diferencia del propuesto, no pueden ser utilizados en la evaluación previa de los pacientes. Con todo, la Tesis ofrece una visión clara sobre la relación entre el SAHS y el habla, evidenciando el grado de madurez alcanzado por la tecnología del habla en la caracterización y detección del SAHS, poniendo de manifiesto que su uso para la evaluación de los pacientes ya sería posible, y dejando la puerta abierta a futuras investigaciones que continúen el trabajo aquí iniciado. ABSTRACT This Thesis explores the potential of speech technologies for the detection of clinical disorders connected to the upper airway. The study of speech traditionally covers both the production process and post processing of the signals involved, from the speaker up to the listener, offering an alternative path to study these pathologies. The fact that utterances embed not just the encoded message but also information about the speaker, has motivated the development of automatic systems oriented to the identification and verificaton the speaker’s identity. These have recently been boosted and reoriented either towards the characterization of traits that are common to several speakers, or to the differences between records of the same speaker collected under different conditions. The first are particularly relevant to this Thesis as these patterns could reveal the presence of features that are related to a common condition shared among different speakers, regardless of their identity. Such is the case faced in this Thesis, where the traits identified would relate to a particular pathology, directly connected to the speech production system. The Obstructive Sleep Apnea syndrome (OSA) is a paradigmatic case for analysis. It is a disorder with high prevalence among adults and affecting a larger number of them as they grow older. Patients suffering from this disorder experience episodes of involuntary cessation of breath during sleep that may last a few seconds and reproduce throughout the night, preventing proper rest. In the case of obstructive apnea, these episodes are related to the collapse of the pharynx, which interrupts the air flow. Currently, OSA diagnosis is done through a polysomnographic study, which focuses on the analysis of apnea episodes during sleep, requiring the patient to stay at the hospital for the whole night. The complexity and high cost of the procedures involved, combined with the waiting lists, have evidenced the need for screening techniques, which perhaps would not achieve outstanding performance rates but would allow clinicians to reorganize these lists ranking patients according to the severity of their condition. Among others, imaging diagnosis and anthropometric characterization of patients have evidenced the existence of anatomical patterns related to OSA that have direct influence on speech. Contributions devoted to the study of how this disorder affects scpeech are scarce and somehow contradictory. However, since the late 1980s the existence of specific patterns related to articulation, phonation and resonance is known. By that time these descriptions were virtually useless when coming to the development of an automatic system, but pointed out the existence of a link between speech and OSA. In recent years automatic processing techniques have evolved and are now able to identify significant differences in the speech of OSAS patients when compared to records from healthy subjects. Nevertheless, little is known about the connection between these new results with those published in the past and the pathogenesis of the OSA syndrome. This Thesis is aimed to progress beyond the previous research done in this area by addressing: the study of how OSA affects patients’ speech, the enhancement of automatic OSA classification based on speech analysis, and its integration with the information embedded in the predictors generally used by clinicians in preliminary patients’ examination. The first two tasks, though may appear symbiotic at first, are quite different. While studying the connection between speech and OSA requires simple narrow models that can be easily interpreted, classification requires larger models including a large number dimensions for the characterization and posterior identification of the observed patterns. Anyhow, it is clear that any progress made in the first task should allow us to improve our performance on the second one, and that the incorporation of the predictors used by clinicians shall contribute in this same direction. The Thesis considers both continuous and sustained speech analysis, to exploit the synergies and differences between them. On continuous speech analysis, a conventional speech processing scheme, designed and evaluated before this Thesis, was taken as a baseline. Over this initial system several alternative representations of the speech information were proposed, optimized and tested to select those more suitable for the characterization of OSA-specific patterns. Evidences were found on the existence of a connection between OSA and the fundamental constituents of the speech: the formants. Experimental results proved that the success of the proposed solution is well explained by the ability of speech representations to describe these specific OSA-related components, ignoring the noisy ones as well those presenting low discrimination capabilities. The resulting scheme obtained a 18% error rate, on a classification scheme significantly less complex than those described in the literature and operating on a single speech record. Regarding the connection between OSA and the observed patterns, it was necessary to consider inter-and intra-group differences for this analysis, and to focus on the articulation, replacing the complex classification models by the long-term average spectra. Results clearly point to certain regions on the frequency axis, suggesting the existence of a systematic narrowing in the vocal tract section at the oropharynx. This was already described in the pathogenesis of this syndrome. Regarding sustained speech, similar experiments as those conducted on continuous speech were reproduced on sustained phonations of vowel / a /. Results were qualitatively similar to the previous ones, though in this case perfomance rates were found to be noticeably lower. Trying to derive further knowledge from this result, experiments on the long-term average spectra and intraand inter-group variability ratios were also reproduced on sustained speech records. Results on both experiments showed significant differences from the previous ones obtained from continuous speech which could explain the differences observed on peformance. However, sustained speech also provided the opportunity to study phonation within the controlled framework it provides. This was also identified in the literature as a source of information for the detection of OSA. In this study it was found that, for the available dataset, no sistematic differences related to phonation could be found between the two groups of speakers. Only those dimensions which relate energy distribution along the frequency axis provided significant differences, pointing once again towards the direction of resonant components. Once classification schemes on both continuous and sustained speech were developed, the Thesis addressed their combination into a single classification system. Under the assumption that the information in continuous and sustained speech is fundamentally different, it should be possible to successfully merge the two of them. This was tested through a simple fusion scheme which obtained a 88.6% correct classification (11.4% error rate), which represents a significant improvement over the state of the art. Finally, the combination of this classifier with the variables used by clinicians obtained a 91.3% accuracy (8.7% error rate). This is within the range of alternative, but costly and intrusive schemes, which unlike the one proposed can not be used in the preliminary assessment of patients’ condition. In the end, this Thesis has shed new light on the underlying connection between OSA and speech, and evidenced the degree of maturity reached by speech technology on OSA characterization and detection, leaving the door open for future research which shall continue in the multiple directions that have been pointed out and left as future work.

Analyzing training dependencies and posterior fusion in discriminant classification of apnoea patients based on sustained and connected speech

We present a novel approach using both sustained vowels and connected speech, to detect obstructive sleep apnea (OSA) cases within a homogeneous group of speakers. The proposed scheme is based on state-of-the-art GMM-based classifiers, and acknowledges specifically the way in which acoustic models are trained on standard databases, as well as the complexity of the resulting models and their adaptation to specific data. Our experimental database contains a suitable number of utterances and sustained speech from healthy (i.e control) and OSA Spanish speakers. Finally, a 25.1% relative reduction in classification error is achieved when fusing continuous and sustained speech classifiers. Index Terms: obstructive sleep apnea (OSA), gaussian mixture models (GMMs), background model (BM), classifier fusion.

On the Mahalanobis Distance Classification Criterion for Multidimensional Normal Distributions

Many existing engineering works model the statistical characteristics of the entities under study as normal distributions. These models are eventually used for decision making, requiring in practice the definition of the classification region corresponding to the desired confidence level. Surprisingly enough, however, a great amount of computer vision works using multidimensional normal models leave unspecified or fail to establish correct confidence regions due to misconceptions on the features of Gaussian functions or to wrong analogies with the unidimensional case. The resulting regions incur in deviations that can be unacceptable in high-dimensional models. Here we provide a comprehensive derivation of the optimal confidence regions for multivariate normal distributions of arbitrary dimensionality. To this end, firstly we derive the condition for region optimality of general continuous multidimensional distributions, and then we apply it to the widespread case of the normal probability density function. The obtained results are used to analyze the confidence error incurred by previous works related to vision research, showing that deviations caused by wrong regions may turn into unacceptable as dimensionality increases. To support the theoretical analysis, a quantitative example in the context of moving object detection by means of background modeling is given.

Improving the management of water and nitrogen resources in cropping systems by using simulation models

Los modelos de simulación de cultivos permiten analizar varias combinaciones de laboreo-rotación y explorar escenarios de manejo. El modelo DSSAT fue evaluado bajo condiciones de secano en un experimento de campo de 16 años en la semiárida España central. Se evaluó el efecto del sistema de laboreo y las rotaciones basadas en cereales de invierno, en el rendimiento del cultivo y la calidad del suelo. Los modelos CERES y CROPGRO se utilizaron para simular el crecimiento y rendimiento del cultivo, mientras que el modelo DSSAT CENTURY se utilizó en las simulaciones de SOC y SN. Tanto las observaciones de campo como las simulaciones con CERES-Barley, mostraron que el rendimiento en grano de la cebada era mas bajo para el cereal continuo (BB) que para las rotaciones de veza (VB) y barbecho (FB) en ambos sistemas de laboreo. El modelo predijo más nitrógeno disponible en el laboreo convencional (CT) que en el no laboreo (NT) conduciendo a un mayor rendimiento en el CT. El SOC y el SN en la capa superficial del suelo, fueron mayores en NT que en CT, y disminuyeron con la profundidad en los valores tanto observados como simulados. Las mejores combinaciones para las condiciones de secano estudiadas fueron CT-VB y CT-FB, pero CT presentó menor contenido en SN y SOC que NT. El efecto beneficioso del NT en SOC y SN bajo condiciones Mediterráneas semiáridas puede ser identificado por observaciones de campo y por simulaciones de modelos de cultivos. La simulación del balance de agua en sistemas de cultivo es una herramienta útil para estudiar como el agua puede ser utilizado eficientemente. La comparación del balance de agua de DSSAT , con una simple aproximación “tipping bucket”, con el modelo WAVE más mecanicista, el cual integra la ecuación de Richard , es un potente método para valorar el funcionamiento del modelo. Los parámetros de suelo fueron calibrados usando el método de optimización global Simulated Annealing (SA). Un lisímetro continuo de pesada en suelo desnudo suministró los valores observados de drenaje y evapotranspiración (ET) mientras que el contenido de agua en el suelo (SW) fue suministrado por sensores de capacitancia. Ambos modelos funcionaron bien después de la optimización de los parámetros de suelo con SA, simulando el balance de agua en el suelo para el período de calibración. Para el período de validación, los modelos optimizados predijeron bien el contenido de agua en el suelo y la evaporación del suelo a lo largo del tiempo. Sin embargo, el drenaje fue predicho mejor con WAVE que con DSSAT, el cual presentó mayores errores en los valores acumulados. Esto podría ser debido a la naturaleza mecanicista de WAVE frente a la naturaleza más funcional de DSSAT. Los buenos resultados de WAVE indican que, después de la calibración, este puede ser utilizado como "benchmark" para otros modelos para periodos en los que no haya medidas de campo del drenaje. El funcionamiento de DSSAT-CENTURY en la simulación de SOC y N depende fuertemente del proceso de inicialización. Se propuso como método alternativo (Met.2) la inicialización de las fracciones de SOC a partir de medidas de mineralización aparente del suelo (Napmin). El Met.2 se comparó con el método de inicialización de Basso et al. (2011) (Met.1), aplicando ambos métodos a un experimento de campo de 4 años en un área en regadío de España central. Nmin y Napmin fueron sobreestimados con el Met.1, ya que la fracción estable obtenida (SOC3) en las capas superficiales del suelo fue más baja que con Met.2. El N lixiviado simulado fue similar en los dos métodos, con buenos resultados en los tratamientos de barbecho y cebada. El Met.1 subestimó el SOC en la capa superficial del suelo cuando se comparó con una serie observada de 12 años. El crecimiento y rendimiento del cultivo fueron adecuadamente simulados con ambos métodos, pero el N en la parte aérea de la planta y en el grano fueron sobreestimados con el Met.1. Los resultados variaron significativamente con las fracciones iniciales de SOC, resaltando la importancia del método de inicialización. El Met.2 ofrece una alternativa para la inicialización del modelo CENTURY, mejorando la simulación de procesos de N en el suelo. La continua emergencia de nuevas variedades de híbridos modernos de maíz limita la aplicación de modelos de simulación de cultivos, ya que estos nuevos híbridos necesitan ser calibrados en el campo para ser adecuados para su uso en los modelos. El desarrollo de relaciones basadas en la duración del ciclo, simplificaría los requerimientos de calibración facilitando la rápida incorporación de nuevos cultivares en DSSAT. Seis híbridos de maiz (FAO 300 hasta FAO 700) fueron cultivados en un experimento de campo de dos años en un área semiárida de regadío en España central. Los coeficientes genéticos fueron obtenidos secuencialmente, comenzando con los parámetros de desarrollo fenológico (P1, P2, P5 and PHINT), seguido de los parámetros de crecimiento del cultivo (G2 and G3). Se continuó el procedimiento hasta que la salida de las simulaciones estuvo en concordancia con las observaciones fenológicas de campo. Después de la calibración, los parámetros simulados se ajustaron bien a los parámetros observados, con bajos RMSE en todos los casos. Los P1 y P5 calibrados, incrementaron con la duración del ciclo. P1 fue una función lineal del tiempo térmico (TT) desde emergencia hasta floración y P5 estuvo linealmente relacionada con el TT desde floración a madurez. No hubo diferencias significativas en PHINT entre híbridos de FAO-500 a 700 , ya que tuvieron un número de hojas similar. Como los coeficientes fenológicos estuvieron directamente relacionados con la duración del ciclo, sería posible desarrollar rangos y correlaciones que permitan estimar dichos coeficientes a partir de la clasificación del ciclo. ABSTRACT Crop simulation models allow analyzing various tillage-rotation combinations and exploring management scenarios. DSSAT model was tested under rainfed conditions in a 16-year field experiment in semiarid central Spain. The effect of tillage system and winter cereal-based rotations on the crop yield and soil quality was evaluated. The CERES and CROPGRO models were used to simulate crop growth and yield, while the DSSAT CENTURY was used in the SOC and SN simulations. Both field observations and CERES-Barley simulations, showed that barley grain yield was lower for continuous cereal (BB) than for vetch (VB) and fallow (FB) rotations for both tillage systems. The model predicted higher nitrogen availability in the conventional tillage (CT) than in the no tillage (NT) leading to a higher yield in the CT. The SOC and SN in the top layer, were higher in NT than in CT, and decreased with depth in both simulated and observed values. The best combinations for the dry land conditions studied were CT-VB and CT-FB, but CT presented lower SN and SOC content than NT. The beneficial effect of NT on SOC and SN under semiarid Mediterranean conditions can be identified by field observations and by crop model simulations. The simulation of the water balance in cropping systems is a useful tool to study how water can be used efficiently. The comparison of DSSAT soil water balance, with a simpler “tipping bucket” approach, with the more mechanistic WAVE model, which integrates Richard’s equation, is a powerful method to assess model performance. The soil parameters were calibrated by using the Simulated Annealing (SA) global optimizing method. A continuous weighing lysimeter in a bare fallow provided the observed values of drainage and evapotranspiration (ET) while soil water content (SW) was supplied by capacitance sensors. Both models performed well after optimizing soil parameters with SA, simulating the soil water balance components for the calibrated period. For the validation period, the optimized models predicted well soil water content and soil evaporation over time. However, drainage was predicted better by WAVE than by DSSAT, which presented larger errors in the cumulative values. That could be due to the mechanistic nature of WAVE against the more functional nature of DSSAT. The good results from WAVE indicate that, after calibration, it could be used as benchmark for other models for periods when no drainage field measurements are available. The performance of DSSAT-CENTURY when simulating SOC and N strongly depends on the initialization process. Initialization of the SOC pools from apparent soil N mineralization (Napmin) measurements was proposed as alternative method (Met.2). Method 2 was compared to the Basso et al. (2011) initialization method (Met.1), by applying both methods to a 4-year field experiment in a irrigated area of central Spain. Nmin and Napmin were overestimated by Met.1, since the obtained stable pool (SOC3) in the upper layers was lower than from Met.2. Simulated N leaching was similar for both methods, with good results in fallow and barley treatments. Method 1 underestimated topsoil SOC when compared with a 12-year observed serial. Crop growth and yield were properly simulated by both methods, but N in shoots and grain were overestimated by Met.1. Results varied significantly with the initial SOC pools, highlighting the importance of the initialization procedure. Method 2 offers an alternative to initialize the CENTURY model, enhancing the simulation of soil N processes. The continuous emergence of new varieties of modern maize hybrids limits the application of crop simulation models, since these new hybrids should be calibrated in the field to be suitable for model use. The development of relationships based on the cycle duration, would simplify the calibration requirements facilitating the rapid incorporation of new cultivars into DSSAT. Six maize hybrids (FAO 300 through FAO 700) were grown in a 2-year field experiment in a semiarid irrigated area of central Spain. Genetic coefficients were obtained sequentially, starting with the phenological development parameters (P1, P2, P5 and PHINT), followed by the crop growth parameters (G2 and G3). The procedure was continued until the simulated outputs were in good agreement with the field phenological observations. After calibration, simulated parameters matched observed parameters well, with low RMSE in most cases. The calibrated P1 and P5 increased with the duration of the cycle. P1 was a linear function of the thermal time (TT) from emergence to silking and P5 was linearly related with the TT from silking to maturity . There were no significant differences in PHINT between hybrids from FAO-500 to 700 , as they had similar leaf number. Since phenological coefficients were directly related with the cycle duration, it would be possible to develop ranges and correlations which allow to estimate such coefficients from the cycle classification.

Expressive power of binary relevance and chain classifiers based on Bayesian Networks for multi-label classification

Bayesian network classifiers are widely used in machine learning because they intuitively represent causal relations. Multi-label classification problems require each instance to be assigned a subset of a defined set of h labels. This problem is equivalent to finding a multi-valued decision function that predicts a vector of h binary classes. In this paper we obtain the decision boundaries of two widely used Bayesian network approaches for building multi-label classifiers: Multi-label Bayesian network classifiers built using the binary relevance method and Bayesian network chain classifiers. We extend our previous single-label results to multi-label chain classifiers, and we prove that, as expected, chain classifiers provide a more expressive model than the binary relevance method.

Analysing the Impact of Consumer Tendency in HORECA Distribution Models

With increasing complexity of today's consumer requirements, food industry decision makers should be able to respond to consumer needs much faster than ever before. The preliminary studies showed that for improving the performance and selecting suitable distribution models, decision makers in food industries should classify different types of consumers and based on the classification prepare different distributions flows. By studying the HORECA distribution channel, this paper suggest that, logistics decision makers should investigate the relationship between consumers' characteristics and urban freight distribution strategy in order to respond to the exact needs and in the follow to reduce the logistics cost.

Bayesian network modeling of the consensus between experts: an application to neuron classification

Neuronal morphology is hugely variable across brain regions and species, and their classification strategies are a matter of intense debate in neuroscience. GABAergic cortical interneurons have been a challenge because it is difficult to find a set of morphological properties which clearly define neuronal types. A group of 48 neuroscience experts around the world were asked to classify a set of 320 cortical GABAergic interneurons according to the main features of their three-dimensional morphological reconstructions. A methodology for building a model which captures the opinions of all the experts was proposed. First, one Bayesian network was learned for each expert, and we proposed an algorithm for clustering Bayesian networks corresponding to experts with similar behaviors. Then, a Bayesian network which represents the opinions of each group of experts was induced. Finally, a consensus Bayesian multinet which models the opinions of the whole group of experts was built. A thorough analysis of the consensus model identified different behaviors between the experts when classifying the interneurons in the experiment. A set of characterizing morphological traits for the neuronal types was defined by performing inference in the Bayesian multinet. These findings were used to validate the model and to gain some insights into neuron morphology.

Spectral study and classification of worldwide locations considering several multijunction solar cell technologies

Multi-junction solar cells are widely used in high-concentration photovoltaic systems (HCPV) attaining the highest efficiencies in photovoltaic energy generation. This technology is more dependent on the spectral variations of the impinging Direct Normal Irradiance (DNI) than conventional photovoltaics based on silicon solar cells and consequently demands a deeper knowledge of the solar resource characteristics. This article explores the capabilities of spectral indexes, namely, spectral matching ratios (SMR), to spectrally characterize the annual irradiation reaching a particular location on the Earth and to provide the necessary information for the spectral optimization of a MJ solar cell in that location as a starting point for CPV module spectral tuning. Additionally, the relationship between such indexes and the atmosphere parameters, such as the aerosol optical depth (AOD), precipitable water (PW), and air mass (AM), is discussed using radiative transfer models such as SMARTS to generate the spectrally-resolved DNI. The network of ground-based sun and sky-scanning radiometers AERONET (AErosol RObotic NETwork) is exploited to obtain the atmosphere parameters for a selected bunch of 34 sites worldwide. Finally, the SMR indexes are obtained for every location, and a comparative analysis is carried out for four architectures of triple junction solar cells, covering both lattice match and metamorphic technologies. The differences found among cell technologies are much less significant than among locations.