Efficient Monte Carlo optimization for multi-label classifier chains

Multi-label classification (MLC) is the supervised learning problem where an instance may be associated with multiple labels. Modeling dependencies between labels allows MLC methods to improve their performance at the expense of an increased computational cost. In this paper we focus on the classifier chains (CC) approach for modeling dependencies. On the one hand, the original CC algorithm makes a greedy approximation, and is fast but tends to propagate errors down the chain. On the other hand, a recent Bayes-optimal method improves the performance, but is computationally intractable in practice. Here we present a novel double-Monte Carlo scheme (M2CC), both for finding a good chain sequence and performing efficient inference. The M2CC algorithm remains tractable for high-dimensional data sets and obtains the best overall accuracy, as shown on several real data sets with input dimension as high as 1449 and up to 103 labels.

A new label fusion method using graph cuts: application to hippocampus segmentation

The aim of this paper is to develop a probabilistic modeling framework for the segmentation of structures of interest from a collection of atlases. Given a subset of registered atlases into the target image for a particular Region of Interest (ROI), a statistical model of appearance and shape is computed for fusing the labels. Segmentations are obtained by minimizing an energy function associated with the proposed model, using a graph-cut technique. We test different label fusion methods on publicly available MR images of human brains.

Análisis de características estáticas de ficheros ejecutables para la clasificación de Malware

El Malware es una grave amenaza para la seguridad de los sistemas. Con el uso generalizado de la World Wide Web, ha habido un enorme aumento en los ataques de virus, haciendo que la seguridad informática sea esencial para todas las computadoras y se expandan las áreas de investigación sobre los nuevos incidentes que se generan, siendo una de éstas la clasificación del malware. Los “desarrolladores de malware” utilizan nuevas técnicas para generar malware polimórfico reutilizando los malware existentes, por lo cual es necesario agruparlos en familias para estudiar sus características y poder detectar nuevas variantes de los mismos. Este trabajo, además de presentar un detallado estado de la cuestión de la clasificación del malware de ficheros ejecutables PE, presenta un enfoque en el que se mejora el índice de la clasificación de la base de datos de Malware MALICIA utilizando las características estáticas de ficheros ejecutables Imphash y Pehash, utilizando dichas características se realiza un clustering con el algoritmo clustering agresivo el cual se cambia con la clasificación actual mediante el algoritmo de majority voting y la característica icon_label, obteniendo un Precision de 99,15% y un Recall de 99,32% mejorando la clasificación de MALICIA con un F-measure de 99,23%.---ABSTRACT---Malware is a serious threat to the security of systems. With the widespread use of the World Wide Web, there has been a huge increase in virus attacks, making the computer security essential for all computers. Near areas of research have append in this area including classifying malware into families, Malware developers use polymorphism to generate new variants of existing malware. Thus it is crucial to group variants of the same family, to study their characteristics and to detect new variants. This work, in addition to presenting a detailed analysis of the problem of classifying malware PE executable files, presents an approach in which the classification in the Malware database MALICIA is improved by using static characteristics of executable files, namely Imphash and Pehash. Both features are evaluated through clustering real malware with family labels with aggressive clustering algorithm and combining this with the current classification by Majority voting algorithm, obtaining a Precision of 99.15% and a Recall of 99.32%, improving the classification of MALICIA with an F-measure of 99,23%.

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.

Multi-dimensional classification of GABAergic interneurons with Bayesian network-modeled label uncertainty

Abstract Interneuron classification is an important and long-debated topic in neuroscience. A recent study provided a data set of digitally reconstructed interneurons classified by 42 leading neuroscientists according to a pragmatic classification scheme composed of five categorical variables, namely, of the interneuron type and four features of axonal morphology. From this data set we now learned a model which can classify interneurons, on the basis of their axonal morphometric parameters, into these five descriptive variables simultaneously. Because of differences in opinion among the neuroscientists, especially regarding neuronal type, for many interneurons we lacked a unique, agreed-upon classification, which we could use to guide model learning. Instead, we guided model learning with a probability distribution over the neuronal type and the axonal features, obtained, for each interneuron, from the neuroscientists’ classification choices. We conveniently encoded such probability distributions with Bayesian networks, calling them label Bayesian networks (LBNs), and developed a method to predict them. This method predicts an LBN by forming a probabilistic consensus among the LBNs of the interneurons most similar to the one being classified. We used 18 axonal morphometric parameters as predictor variables, 13 of which we introduce in this paper as quantitative counterparts to the categorical axonal features. We were able to accurately predict interneuronal LBNs. Furthermore, when extracting crisp (i.e., non-probabilistic) predictions from the predicted LBNs, our method outperformed related work on interneuron classification. Our results indicate that our method is adequate for multi-dimensional classification of interneurons with probabilistic labels. Moreover, the introduced morphometric parameters are good predictors of interneuron type and the four features of axonal morphology and thus may serve as objective counterparts to the subjective, categorical axonal features.

Multi-dimensional classification of GABAergic interneurons with Bayesian network-modeled label uncertainty

Interneuron classification is an important and long-debated topic in neuroscience. A recent study provided a data set of digitally reconstructed interneurons classified by 42 leading neuroscientists according to a pragmatic classification scheme composed of five categorical variables, namely, of the interneuron type and four features of axonal morphology. From this data set we now learned a model which can classify interneurons, on the basis of their axonal morphometric parameters, into these five descriptive variables simultaneously. Because of differences in opinion among the neuroscientists, especially regarding neuronal type, for many interneurons we lacked a unique, agreed-upon classification, which we could use to guide model learning. Instead, we guided model learning with a probability distribution over the neuronal type and the axonal features, obtained, for each interneuron, from the neuroscientists’ classification choices. We conveniently encoded such probability distributions with Bayesian networks, calling them label Bayesian networks (LBNs), and developed a method to predict them. This method predicts an LBN by forming a probabilistic consensus among the LBNs of the interneurons most similar to the one being classified. We used 18 axonal morphometric parameters as predictor variables, 13 of which we introduce in this paper as quantitative counterparts to the categorical axonal features. We were able to accurately predict interneuronal LBNs. Furthermore, when extracting crisp (i.e., non-probabilistic) predictions from the predicted LBNs, our method outperformed related work on interneuron classification. Our results indicate that our method is adequate for multi-dimensional classification of interneurons with probabilistic labels. Moreover, the introduced morphometric parameters are good predictors of interneuron type and the four features of axonal morphology and thus may serve as objective counterparts to the subjective, categorical axonal features.

Classifying GABAergic interneurons with semi-supervised projected model-based clustering

Objectives: A recently introduced pragmatic scheme promises to be a useful catalog of interneuron names.We sought to automatically classify digitally reconstructed interneuronal morphologies according tothis scheme. Simultaneously, we sought to discover possible subtypes of these types that might emergeduring automatic classification (clustering). We also investigated which morphometric properties weremost relevant for this classification.Materials and methods: A set of 118 digitally reconstructed interneuronal morphologies classified into thecommon basket (CB), horse-tail (HT), large basket (LB), and Martinotti (MA) interneuron types by 42 of theworld?s leading neuroscientists, quantified by five simple morphometric properties of the axon and fourof the dendrites. We labeled each neuron with the type most commonly assigned to it by the experts. Wethen removed this class information for each type separately, and applied semi-supervised clustering tothose cells (keeping the others? cluster membership fixed), to assess separation from other types and lookfor the formation of new groups (subtypes). We performed this same experiment unlabeling the cells oftwo types at a time, and of half the cells of a single type at a time. The clustering model is a finite mixtureof Gaussians which we adapted for the estimation of local (per-cluster) feature relevance. We performedthe described experiments on three different subsets of the data, formed according to how many expertsagreed on type membership: at least 18 experts (the full data set), at least 21 (73 neurons), and at least26 (47 neurons).Results: Interneurons with more reliable type labels were classified more accurately. We classified HTcells with 100% accuracy, MA cells with 73% accuracy, and CB and LB cells with 56% and 58% accuracy,respectively. We identified three subtypes of the MA type, one subtype of CB and LB types each, andno subtypes of HT (it was a single, homogeneous type). We got maximum (adapted) Silhouette widthand ARI values of 1, 0.83, 0.79, and 0.42, when unlabeling the HT, CB, LB, and MA types, respectively,confirming the quality of the formed cluster solutions. The subtypes identified when unlabeling a singletype also emerged when unlabeling two types at a time, confirming their validity. Axonal morphometricproperties were more relevant that dendritic ones, with the axonal polar histogram length in the [pi, 2pi) angle interval being particularly useful.Conclusions: The applied semi-supervised clustering method can accurately discriminate among CB, HT, LB, and MA interneuron types while discovering potential subtypes, and is therefore useful for neuronal classification. The discovery of potential subtypes suggests that some of these types are more heteroge-neous that previously thought. Finally, axonal variables seem to be more relevant than dendritic ones fordistinguishing among the CB, HT, LB, and MA interneuron types.

Efficient Monte Carlo methods for multi-dimensional learning with classifier chains

Multi-dimensional classification (MDC) is the supervised learning problem where an instance is associated with multiple classes, rather than with a single class, as in traditional classification problems. Since these classes are often strongly correlated, modeling the dependencies between them allows MDC methods to improve their performance – at the expense of an increased computational cost. In this paper we focus on the classifier chains (CC) approach for modeling dependencies, one of the most popular and highest-performing methods for multi-label classification (MLC), a particular case of MDC which involves only binary classes (i.e., labels). The original CC algorithm makes a greedy approximation, and is fast but tends to propagate errors along the chain. Here we present novel Monte Carlo schemes, both for finding a good chain sequence and performing efficient inference. Our algorithms remain tractable for high-dimensional data sets and obtain the best predictive performance across several real data sets.

Neuronanatomy, neurology and Bayesian networks

Bayesian networks are data mining models with clear semantics and a sound theoretical foundation. In this keynote talk we will pinpoint a number of neuroscience problems that can be addressed using Bayesian networks. In neuroanatomy, we will show computer simulation models of dendritic trees and classification of neuron types, both based on morphological features. In neurology, we will present the search for genetic biomarkers in Alzheimer's disease and the prediction of health-related quality of life in Parkinson's disease. Most of these challenging problems posed by neuroscience involve new Bayesian network designs that can cope with multiple class variables, small sample sizes, or labels annotated by several experts.

Using Open Geographic Data to Generate Natural Language Descriptions for Hydrological Sensor Networks

Providing descriptions of isolated sensors and sensor networks in natural language, understandable by the general public, is useful to help users find relevant sensors and analyze sensor data. In this paper, we discuss the feasibility of using geographic knowledge from public databases available on the Web (such as OpenStreetMap, Geonames, or DBpedia) to automatically construct such descriptions. We present a general method that uses such information to generate sensor descriptions in natural language. The results of the evaluation of our method in a hydrologic national sensor network showed that this approach is feasible and capable of generating adequate sensor descriptions with a lower development effort compared to other approaches. In the paper we also analyze certain problems that we found in public databases (e.g., heterogeneity, non-standard use of labels, or rigid search methods) and their impact in the generation of sensor descriptions.

Dynamic topic-based adaptation of language models: a comparison between different approaches

This paper presents a dynamic LM adaptation based on the topic that has been identified on a speech segment. We use LSA and the given topic labels in the training dataset to obtain and use the topic models. We propose a dynamic language model adaptation to improve the recognition performance in "a two stages" AST system. The final stage makes use of the topic identification with two variants: the first on uses just the most probable topic and the other one depends on the relative distances of the topics that have been identified. We perform the adaptation of the LM as a linear interpolation between a background model and topic-based LM. The interpolation weight id dynamically adapted according to different parameters. The proposed method is evaluated on the Spanish partition of the EPPS speech database. We achieved a relative reduction in WER of 11.13% over the baseline system which uses a single blackground LM.

New definitions of margin for multi-class Boosting algorithms

La familia de algoritmos de Boosting son un tipo de técnicas de clasificación y regresión que han demostrado ser muy eficaces en problemas de Visión Computacional. Tal es el caso de los problemas de detección, de seguimiento o bien de reconocimiento de caras, personas, objetos deformables y acciones. El primer y más popular algoritmo de Boosting, AdaBoost, fue concebido para problemas binarios. Desde entonces, muchas han sido las propuestas que han aparecido con objeto de trasladarlo a otros dominios más generales: multiclase, multilabel, con costes, etc. Nuestro interés se centra en extender AdaBoost al terreno de la clasificación multiclase, considerándolo como un primer paso para posteriores ampliaciones. En la presente tesis proponemos dos algoritmos de Boosting para problemas multiclase basados en nuevas derivaciones del concepto margen. El primero de ellos, PIBoost, está concebido para abordar el problema descomponiéndolo en subproblemas binarios. Por un lado, usamos una codificación vectorial para representar etiquetas y, por otro, utilizamos la función de pérdida exponencial multiclase para evaluar las respuestas. Esta codificación produce un conjunto de valores margen que conllevan un rango de penalizaciones en caso de fallo y recompensas en caso de acierto. La optimización iterativa del modelo genera un proceso de Boosting asimétrico cuyos costes dependen del número de etiquetas separadas por cada clasificador débil. De este modo nuestro algoritmo de Boosting tiene en cuenta el desbalanceo debido a las clases a la hora de construir el clasificador. El resultado es un método bien fundamentado que extiende de manera canónica al AdaBoost original. El segundo algoritmo propuesto, BAdaCost, está concebido para problemas multiclase dotados de una matriz de costes. Motivados por los escasos trabajos dedicados a generalizar AdaBoost al terreno multiclase con costes, hemos propuesto un nuevo concepto de margen que, a su vez, permite derivar una función de pérdida adecuada para evaluar costes. Consideramos nuestro algoritmo como la extensión más canónica de AdaBoost para este tipo de problemas, ya que generaliza a los algoritmos SAMME, Cost-Sensitive AdaBoost y PIBoost. Por otro lado, sugerimos un simple procedimiento para calcular matrices de coste adecuadas para mejorar el rendimiento de Boosting a la hora de abordar problemas estándar y problemas con datos desbalanceados. Una serie de experimentos nos sirven para demostrar la efectividad de ambos métodos frente a otros conocidos algoritmos de Boosting multiclase en sus respectivas áreas. En dichos experimentos se usan bases de datos de referencia en el área de Machine Learning, en primer lugar para minimizar errores y en segundo lugar para minimizar costes. Además, hemos podido aplicar BAdaCost con éxito a un proceso de segmentación, un caso particular de problema con datos desbalanceados. Concluimos justificando el horizonte de futuro que encierra el marco de trabajo que presentamos, tanto por su aplicabilidad como por su flexibilidad teórica. Abstract The family of Boosting algorithms represents a type of classification and regression approach that has shown to be very effective in Computer Vision problems. Such is the case of detection, tracking and recognition of faces, people, deformable objects and actions. The first and most popular algorithm, AdaBoost, was introduced in the context of binary classification. Since then, many works have been proposed to extend it to the more general multi-class, multi-label, costsensitive, etc... domains. Our interest is centered in extending AdaBoost to two problems in the multi-class field, considering it a first step for upcoming generalizations. In this dissertation we propose two Boosting algorithms for multi-class classification based on new generalizations of the concept of margin. The first of them, PIBoost, is conceived to tackle the multi-class problem by solving many binary sub-problems. We use a vectorial codification to represent class labels and a multi-class exponential loss function to evaluate classifier responses. This representation produces a set of margin values that provide a range of penalties for failures and rewards for successes. The stagewise optimization of this model introduces an asymmetric Boosting procedure whose costs depend on the number of classes separated by each weak-learner. In this way the Boosting procedure takes into account class imbalances when building the ensemble. The resulting algorithm is a well grounded method that canonically extends the original AdaBoost. The second algorithm proposed, BAdaCost, is conceived for multi-class problems endowed with a cost matrix. Motivated by the few cost-sensitive extensions of AdaBoost to the multi-class field, we propose a new margin that, in turn, yields a new loss function appropriate for evaluating costs. Since BAdaCost generalizes SAMME, Cost-Sensitive AdaBoost and PIBoost algorithms, we consider our algorithm as a canonical extension of AdaBoost to this kind of problems. We additionally suggest a simple procedure to compute cost matrices that improve the performance of Boosting in standard and unbalanced problems. A set of experiments is carried out to demonstrate the effectiveness of both methods against other relevant Boosting algorithms in their respective areas. In the experiments we resort to benchmark data sets used in the Machine Learning community, firstly for minimizing classification errors and secondly for minimizing costs. In addition, we successfully applied BAdaCost to a segmentation task, a particular problem in presence of imbalanced data. We conclude the thesis justifying the horizon of future improvements encompassed in our framework, due to its applicability and theoretical flexibility.

Contributions to Speech Analytics based on Speech Recognition and Topic Identification

La última década ha sido testigo de importantes avances en el campo de la tecnología de reconocimiento de voz. Los sistemas comerciales existentes actualmente poseen la capacidad de reconocer habla continua de múltiples locutores, consiguiendo valores aceptables de error, y sin la necesidad de realizar procedimientos explícitos de adaptación. A pesar del buen momento que vive esta tecnología, el reconocimiento de voz dista de ser un problema resuelto. La mayoría de estos sistemas de reconocimiento se ajustan a dominios particulares y su eficacia depende de manera significativa, entre otros muchos aspectos, de la similitud que exista entre el modelo de lenguaje utilizado y la tarea específica para la cual se está empleando. Esta dependencia cobra aún más importancia en aquellos escenarios en los cuales las propiedades estadísticas del lenguaje varían a lo largo del tiempo, como por ejemplo, en dominios de aplicación que involucren habla espontánea y múltiples temáticas. En los últimos años se ha evidenciado un constante esfuerzo por mejorar los sistemas de reconocimiento para tales dominios. Esto se ha hecho, entre otros muchos enfoques, a través de técnicas automáticas de adaptación. Estas técnicas son aplicadas a sistemas ya existentes, dado que exportar el sistema a una nueva tarea o dominio puede requerir tiempo a la vez que resultar costoso. Las técnicas de adaptación requieren fuentes adicionales de información, y en este sentido, el lenguaje hablado puede aportar algunas de ellas. El habla no sólo transmite un mensaje, también transmite información acerca del contexto en el cual se desarrolla la comunicación hablada (e.g. acerca del tema sobre el cual se está hablando). Por tanto, cuando nos comunicamos a través del habla, es posible identificar los elementos del lenguaje que caracterizan el contexto, y al mismo tiempo, rastrear los cambios que ocurren en estos elementos a lo largo del tiempo. Esta información podría ser capturada y aprovechada por medio de técnicas de recuperación de información (information retrieval) y de aprendizaje de máquina (machine learning). Esto podría permitirnos, dentro del desarrollo de mejores sistemas automáticos de reconocimiento de voz, mejorar la adaptación de modelos del lenguaje a las condiciones del contexto, y por tanto, robustecer al sistema de reconocimiento en dominios con condiciones variables (tales como variaciones potenciales en el vocabulario, el estilo y la temática). En este sentido, la principal contribución de esta Tesis es la propuesta y evaluación de un marco de contextualización motivado por el análisis temático y basado en la adaptación dinámica y no supervisada de modelos de lenguaje para el robustecimiento de un sistema automático de reconocimiento de voz. Esta adaptación toma como base distintos enfoque de los sistemas mencionados (de recuperación de información y aprendizaje de máquina) mediante los cuales buscamos identificar las temáticas sobre las cuales se está hablando en una grabación de audio. Dicha identificación, por lo tanto, permite realizar una adaptación del modelo de lenguaje de acuerdo a las condiciones del contexto. El marco de contextualización propuesto se puede dividir en dos sistemas principales: un sistema de identificación de temática y un sistema de adaptación dinámica de modelos de lenguaje. Esta Tesis puede describirse en detalle desde la perspectiva de las contribuciones particulares realizadas en cada uno de los campos que componen el marco propuesto: _ En lo referente al sistema de identificación de temática, nos hemos enfocado en aportar mejoras a las técnicas de pre-procesamiento de documentos, asimismo en contribuir a la definición de criterios más robustos para la selección de index-terms. – La eficiencia de los sistemas basados tanto en técnicas de recuperación de información como en técnicas de aprendizaje de máquina, y específicamente de aquellos sistemas que particularizan en la tarea de identificación de temática, depende, en gran medida, de los mecanismos de preprocesamiento que se aplican a los documentos. Entre las múltiples operaciones que hacen parte de un esquema de preprocesamiento, la selección adecuada de los términos de indexado (index-terms) es crucial para establecer relaciones semánticas y conceptuales entre los términos y los documentos. Este proceso también puede verse afectado, o bien por una mala elección de stopwords, o bien por la falta de precisión en la definición de reglas de lematización. En este sentido, en este trabajo comparamos y evaluamos diferentes criterios para el preprocesamiento de los documentos, así como también distintas estrategias para la selección de los index-terms. Esto nos permite no sólo reducir el tamaño de la estructura de indexación, sino también mejorar el proceso de identificación de temática. – Uno de los aspectos más importantes en cuanto al rendimiento de los sistemas de identificación de temática es la asignación de diferentes pesos a los términos de acuerdo a su contribución al contenido del documento. En este trabajo evaluamos y proponemos enfoques alternativos a los esquemas tradicionales de ponderado de términos (tales como tf-idf ) que nos permitan mejorar la especificidad de los términos, así como también discriminar mejor las temáticas de los documentos. _ Respecto a la adaptación dinámica de modelos de lenguaje, hemos dividimos el proceso de contextualización en varios pasos. – Para la generación de modelos de lenguaje basados en temática, proponemos dos tipos de enfoques: un enfoque supervisado y un enfoque no supervisado. En el primero de ellos nos basamos en las etiquetas de temática que originalmente acompañan a los documentos del corpus que empleamos. A partir de estas, agrupamos los documentos que forman parte de la misma temática y generamos modelos de lenguaje a partir de dichos grupos. Sin embargo, uno de los objetivos que se persigue en esta Tesis es evaluar si el uso de estas etiquetas para la generación de modelos es óptimo en términos del rendimiento del reconocedor. Por esta razón, nosotros proponemos un segundo enfoque, un enfoque no supervisado, en el cual el objetivo es agrupar, automáticamente, los documentos en clusters temáticos, basándonos en la similaridad semántica existente entre los documentos. Por medio de enfoques de agrupamiento conseguimos mejorar la cohesión conceptual y semántica en cada uno de los clusters, lo que a su vez nos permitió refinar los modelos de lenguaje basados en temática y mejorar el rendimiento del sistema de reconocimiento. – Desarrollamos diversas estrategias para generar un modelo de lenguaje dependiente del contexto. Nuestro objetivo es que este modelo refleje el contexto semántico del habla, i.e. las temáticas más relevantes que se están discutiendo. Este modelo es generado por medio de la interpolación lineal entre aquellos modelos de lenguaje basados en temática que estén relacionados con las temáticas más relevantes. La estimación de los pesos de interpolación está basada principalmente en el resultado del proceso de identificación de temática. – Finalmente, proponemos una metodología para la adaptación dinámica de un modelo de lenguaje general. El proceso de adaptación tiene en cuenta no sólo al modelo dependiente del contexto sino también a la información entregada por el proceso de identificación de temática. El esquema usado para la adaptación es una interpolación lineal entre el modelo general y el modelo dependiente de contexto. Estudiamos también diferentes enfoques para determinar los pesos de interpolación entre ambos modelos. Una vez definida la base teórica de nuestro marco de contextualización, proponemos su aplicación dentro de un sistema automático de reconocimiento de voz. Para esto, nos enfocamos en dos aspectos: la contextualización de los modelos de lenguaje empleados por el sistema y la incorporación de información semántica en el proceso de adaptación basado en temática. En esta Tesis proponemos un marco experimental basado en una arquitectura de reconocimiento en ‘dos etapas’. En la primera etapa, empleamos sistemas basados en técnicas de recuperación de información y aprendizaje de máquina para identificar las temáticas sobre las cuales se habla en una transcripción de un segmento de audio. Esta transcripción es generada por el sistema de reconocimiento empleando un modelo de lenguaje general. De acuerdo con la relevancia de las temáticas que han sido identificadas, se lleva a cabo la adaptación dinámica del modelo de lenguaje. En la segunda etapa de la arquitectura de reconocimiento, usamos este modelo adaptado para realizar de nuevo el reconocimiento del segmento de audio. Para determinar los beneficios del marco de trabajo propuesto, llevamos a cabo la evaluación de cada uno de los sistemas principales previamente mencionados. Esta evaluación es realizada sobre discursos en el dominio de la política usando la base de datos EPPS (European Parliamentary Plenary Sessions - Sesiones Plenarias del Parlamento Europeo) del proyecto europeo TC-STAR. Analizamos distintas métricas acerca del rendimiento de los sistemas y evaluamos las mejoras propuestas con respecto a los sistemas de referencia. ABSTRACT The last decade has witnessed major advances in speech recognition technology. Today’s commercial systems are able to recognize continuous speech from numerous speakers, with acceptable levels of error and without the need for an explicit adaptation procedure. Despite this progress, speech recognition is far from being a solved problem. Most of these systems are adjusted to a particular domain and their efficacy depends significantly, among many other aspects, on the similarity between the language model used and the task that is being addressed. This dependence is even more important in scenarios where the statistical properties of the language fluctuates throughout the time, for example, in application domains involving spontaneous and multitopic speech. Over the last years there has been an increasing effort in enhancing the speech recognition systems for such domains. This has been done, among other approaches, by means of techniques of automatic adaptation. These techniques are applied to the existing systems, specially since exporting the system to a new task or domain may be both time-consuming and expensive. Adaptation techniques require additional sources of information, and the spoken language could provide some of them. It must be considered that speech not only conveys a message, it also provides information on the context in which the spoken communication takes place (e.g. on the subject on which it is being talked about). Therefore, when we communicate through speech, it could be feasible to identify the elements of the language that characterize the context, and at the same time, to track the changes that occur in those elements over time. This information can be extracted and exploited through techniques of information retrieval and machine learning. This allows us, within the development of more robust speech recognition systems, to enhance the adaptation of language models to the conditions of the context, thus strengthening the recognition system for domains under changing conditions (such as potential variations in vocabulary, style and topic). In this sense, the main contribution of this Thesis is the proposal and evaluation of a framework of topic-motivated contextualization based on the dynamic and non-supervised adaptation of language models for the enhancement of an automatic speech recognition system. This adaptation is based on an combined approach (from the perspective of both information retrieval and machine learning fields) whereby we identify the topics that are being discussed in an audio recording. The topic identification, therefore, enables the system to perform an adaptation of the language model according to the contextual conditions. The proposed framework can be divided in two major systems: a topic identification system and a dynamic language model adaptation system. This Thesis can be outlined from the perspective of the particular contributions made in each of the fields that composes the proposed framework: _ Regarding the topic identification system, we have focused on the enhancement of the document preprocessing techniques in addition to contributing in the definition of more robust criteria for the selection of index-terms. – Within both information retrieval and machine learning based approaches, the efficiency of topic identification systems, depends, to a large extent, on the mechanisms of preprocessing applied to the documents. Among the many operations that encloses the preprocessing procedures, an adequate selection of index-terms is critical to establish conceptual and semantic relationships between terms and documents. This process might also be weakened by a poor choice of stopwords or lack of precision in defining stemming rules. In this regard we compare and evaluate different criteria for preprocessing the documents, as well as for improving the selection of the index-terms. This allows us to not only reduce the size of the indexing structure but also to strengthen the topic identification process. – One of the most crucial aspects, in relation to the performance of topic identification systems, is to assign different weights to different terms depending on their contribution to the content of the document. In this sense we evaluate and propose alternative approaches to traditional weighting schemes (such as tf-idf ) that allow us to improve the specificity of terms, and to better identify the topics that are related to documents. _ Regarding the dynamic language model adaptation, we divide the contextualization process into different steps. – We propose supervised and unsupervised approaches for the generation of topic-based language models. The first of them is intended to generate topic-based language models by grouping the documents, in the training set, according to the original topic labels of the corpus. Nevertheless, a goal of this Thesis is to evaluate whether or not the use of these labels to generate language models is optimal in terms of recognition accuracy. For this reason, we propose a second approach, an unsupervised one, in which the objective is to group the data in the training set into automatic topic clusters based on the semantic similarity between the documents. By means of clustering approaches we expect to obtain a more cohesive association of the documents that are related by similar concepts, thus improving the coverage of the topic-based language models and enhancing the performance of the recognition system. – We develop various strategies in order to create a context-dependent language model. Our aim is that this model reflects the semantic context of the current utterance, i.e. the most relevant topics that are being discussed. This model is generated by means of a linear interpolation between the topic-based language models related to the most relevant topics. The estimation of the interpolation weights is based mainly on the outcome of the topic identification process. – Finally, we propose a methodology for the dynamic adaptation of a background language model. The adaptation process takes into account the context-dependent model as well as the information provided by the topic identification process. The scheme used for the adaptation is a linear interpolation between the background model and the context-dependent one. We also study different approaches to determine the interpolation weights used in this adaptation scheme. Once we defined the basis of our topic-motivated contextualization framework, we propose its application into an automatic speech recognition system. We focus on two aspects: the contextualization of the language models used by the system, and the incorporation of semantic-related information into a topic-based adaptation process. To achieve this, we propose an experimental framework based in ‘a two stages’ recognition architecture. In the first stage of the architecture, Information Retrieval and Machine Learning techniques are used to identify the topics in a transcription of an audio segment. This transcription is generated by the recognition system using a background language model. According to the confidence on the topics that have been identified, the dynamic language model adaptation is carried out. In the second stage of the recognition architecture, an adapted language model is used to re-decode the utterance. To test the benefits of the proposed framework, we carry out the evaluation of each of the major systems aforementioned. The evaluation is conducted on speeches of political domain using the EPPS (European Parliamentary Plenary Sessions) database from the European TC-STAR project. We analyse several performance metrics that allow us to compare the improvements of the proposed systems against the baseline ones.

Diseño e implementación de un módulo para la estructuración de notas clínicas

El presente Trabajo Fin de Grado (TFG) surge de la necesidad de disponer de tecnologías que faciliten el Procesamiento de Lenguaje Natural (NLP) en español dentro del sector de la medicina. Centrado concretamente en la extracción de conocimiento de las historias clínicas electrónicas (HCE), que recogen toda la información relacionada con la salud del paciente y en particular, de los documentos recogidos en dichas historias, pretende la obtención de todos los términos relacionados con la medicina. El Procesamiento de Lenguaje Natural permite la obtención de datos estructurados a partir de información no estructurada. Estas técnicas permiten un análisis de texto que genera etiquetas aportando significado semántico a las palabras para la manipulación de información. A partir de la investigación realizada del estado del arte en NLP y de las tecnologías existentes para otras lenguas, se propone como solución un módulo de anotación de términos médicos extraídos de documentos clínicos. Como términos médicos se han considerado síntomas, enfermedades, partes del cuerpo o tratamientos obtenidos de UMLS, una ontología categorizada que agrega distintas fuentes de datos médicos. Se ha realizado el diseño y la implementación del módulo así como el análisis de los resultados obtenidos realizando una evaluación con treinta y dos documentos que contenían 1372 menciones de terminología médica y que han dado un resultado medio de Precisión: 70,4%, Recall: 36,2%, Accuracy: 31,4% y F-Measure: 47,2%.---ABSTRACT---This Final Thesis arises from the need for technologies that facilitate the Natural Language Processing (NLP) in Spanish in the medical sector. Specifically it is focused on extracting knowledge from Electronic Health Records (EHR), which contain all the information related to the patient's health and, in particular, it expects to obtain all the terms related to medicine from the documents contained in these records. Natural Language Processing allows us to obtain structured information from unstructured data. These techniques enable analysis of text generating labels providing semantic meaning to words for handling information. From the investigation of the state of the art in NLP and existing technologies in other languages, an annotation module of medical terms extracted from clinical documents is proposed as a solution. Symptoms, diseases, body parts or treatments are considered part of the medical terms contained in UMLS ontology which is categorized joining different sources of medical data. This project has completed the design and implementation of a module and the analysis of the results have been obtained. Thirty two documents which contain 1372 mentions of medical terminology have been evaluated and the average results obtained are: Precision: 70.4% Recall: 36.2% Accuracy: 31.4% and F-Measure: 47.2%.

Definición de los procedimientos e indicadores de la cadena de custodia de la industria de pasta y papel vinculados a la normativa ISO 9000/14000

En el trabajo se ha reflejado la génesis del término „desarrollo sostenible‟ y la preocupación social y de los gobiernos hacia la sostenibilidad de los recursos, generando una política forestal a escala internacional, cuyo hito inicial desempeña la Cumbre de Rio y otros procesos, donde surge el término „sostenibilidad‟ - procesos gubernamentales que ponen su atención en los bosques,entre ellos el Proceso paneuropeo del que emanan los Criterios e indicadores de la gestión forestal sostenible, así como otros foros donde aparece la vinculación entre los productos y la sostenibilidad de los bosques de donde proceden. A partir de esos conceptos aparecen los sistemas de certificación, primero FSC, promovido por las ONG ambientalistas, y en respuesta a ello, el sistema PEFC de los propietarios forestales privados con las industrias forestales. En su introducción, el trabajo recoge como surge y evoluciona esta preocupación y como se sucede y desarrollan los sistemas de certificación en España, orientados tanto a la superficie forestal como a sus productos derivados. Tratando de ser una herramienta técnica, en estos últimos quince años, se han ido planteando metodologías y propuestas técnicas que puedan garantizar el origen sostenible del bosque de donde procede la materia prima. Aunque la persistencia de la masa forestal como término tiene importancia, no considera numerosos aspectos incluidos en el concepto “gestión forestal sostenible”. El trabajo manifiesta qué es y cómo se origina la Gestión Forestal Sostenible y muy ligado con ello, el origen de la certificación como herramienta, explicando cómo se incardina la certificación forestal española con las estructuras internacionales y se construyen nuevas estructuras a escala española y autonómica. Se expone el dominio y alcance técnico de la certificación, sus inicios y un análisis de la diversidad de sistemas y etiquetas existentes -como ha ido evolucionando a escala internacional y nacional, por sistemas, sectores y continentes, con especial atención en España, recopilando lo sucedido relevante para el sector forestal, industrias forestales y sistemas de certificación en los años de estudio de elaboración de la tesis. Se refleja la relevancia que adquiere la certificación forestal y de la cadena de custodia para la industria del papel, tanto por las regulaciones, normativas y legislación que involucran al producto derivado del bosque como por ser herramienta que enlaza el bosque sostenible con el consumidor. La aplicación de todas esas especificaciones técnicas que muestran la sostenibilidad del bosque y al trazabilidad en el proceso productivo comporta una carga administrativa de recopilación de información, de control para el seguimiento asociado con los registros necesarios, y de archivo de documentos, conforme a las exigencias que plantean los sistemas de certificación forestal. Por tanto, es importante definir un método y procedimientos genéricos para los correspondientes sistemas de gestión preexistentes en las empresas del sector de pasta y papel (de calidad/ de medio ambiente/integrados), para implantar un Sistema de Cadena de Custodia genérico (común a FSC y PEFC) en una instalación de celulosa y papel y un sistema de diligencia debida. Para ello, ha sido necesario analizar la línea de producción y establecer los puntos de su diagrama en los que se requiere el control de la trazabilidad, puntos en los que se procede al seguimiento y registro de la materia prima, materia semielaborada y producto, y de ahí proceder a redactar o retocar los procedimientos de gestión calidad/ medioambiental, en su caso, para incluir los campos de registro. Según determinen los procedimientos, se efectuará un seguimiento y registro de los, derivados que configuran una característica y se determinará una serie de indicadores del sistema de trazabilidad de la madera en la industria de celulosa y papel, es decir, un conjunto de parámetros cuantitativos y descriptivos, sujetos al seguimiento de forma periódica, que muestren el cambio y permitan observar la evaluación y control del Sistema de Cadena de Custodia. Además de asegurar la trazabilidad de la madera y fibra en la industria de pasta y papel y con ello la sostenibilidad del bosque del que procede, se avalará la legalidad de los aprovechamientos que proporcionan ese recurso maderable, cumpliendo así no sólo la legislación vigente sino también ambos sistemas de certificación FSC y PEFC. El sistema de cadena de custodia de la pasta y papel se caracteriza por los indicadores de seguimiento que permite el control de la trazabilidad. ABSTRACT This paper traces the origins of the term „Sustainable Development‟ and of both citizen and institutional concern for the sustainability of resources, leading to the enactment of a forestry policy at international level, of which the initial milestones are the Rio Summit and other processes in which the term „Sustainability‟ was born. Those forestfocused institutional initiatives include the pan-European process that led to the publication of Sustainable Forest Management Criteria and Indicators, and other forums that highlight the link between finished wood-based products and the sustainability of the forests from which that wood is sourced. Those concepts were the culture in which forest certification systems were engendered, first FSC, promoted by environmental NGOs, and subsequently PEFC, fostered in response to FSC by private forest owners and forest-based industries. In its Introduction, this paper looks at how such concern arose and has evolved and how certification systems came into existence in Spain and developed to encompass both forest lands and forest-based products. As part of a mission to provide an applicable technical tool, new methodologies and technical proposals have been put forward over the past fifteen years aimed at guaranteeing the sustainable origin of the forest from which raw material is sourced. Although the maintenance of forest stands as a term in its own right is important, it does not take many of the aspects included in the concept of “sustainable forest management” into account. This thesis describes what SFM is and how it was born, underlying the close link between SFM and the origin of certification as a tool, explaining how Spanish forest certification is embodied in international structures, while new structures are built here in Spain on both the national and regional scale. This work also details the domain and scope of forest certification from the technical standpoint, explains its beginnings, and assesses the various systems and labels that exist - how certification has evolved internationally and nationally across systems, sectors and continents, with special emphasis on Spain. It provides a compilation of events relevant to forestry, forest industries and forest certification systems that have taken place during the years this thesis has been in preparation. It reflects on the relevance that forest and chain of custody certification holds for the paper industry, in view not only of the regulations, policies and legislation that affect forest-based products but also of its role as a tool that bonds the sustainable forest with the consumer. Implementing the range of technical specifications to demonstrate forest sustainability and traceability throughout the production process entails the administrative burden of collecting information and providing controls to process the relevant records and documents to be kept on file in compliance with the requirements made by forest certification schemes. It is therefore important to define a generic method, together with its relevant procedures,that fits the management systems (quality / environmental / integrated)existing today in pulp and paper companies, in order to implement a generic Chain of Custody scheme (common to FSC and PEFC) in a pulp and paper mill, and a due diligence system. To achieve that, it has first been necessary to analyse the production line and establish points along the route where traceabilitycontrols need to be implemented and points where raw material, semi-finished goods and end products should be monitored and recorded. Subsequently, the procedures in quality / environmental management systems need to be drafted or amended as required to include fields that reflect those records. As required by the procedures, forest-based products that have a specific characteristic shall be monitored and recorded, and a number of indicators identified in the traceability system of wood for pulp & paper, i.e.createa set of quantitative and descriptive parameters subject to regular monitoringthat reveal changes and enable the Chain of Custody system to be assessed and controlled. In addition to ensuring the traceability of wood and fibre in the pulp and paper industry -and so the sustainability of the forest from which it is sourced -, the legality of the harvesting that produces that timber will also be enhanced, thus fulfilling not only the law but also both FSC and PEFC certification schemes. The chain of custody system for pulp and paper is characterised by monitoring indicators that enable traceability to be controlled.