An Interval-based Multiobjective Approach to Feature Subset Selection Using Joint Modeling of Objectives and Variables

This paper studies feature subset selection in classification using a multiobjective estimation of distribution algorithm. We consider six functions, namely area under ROC curve, sensitivity, specificity, precision, F1 measure and Brier score, for evaluation of feature subsets and as the objectives of the problem. One of the characteristics of these objective functions is the existence of noise in their values that should be appropriately handled during optimization. Our proposed algorithm consists of two major techniques which are specially designed for the feature subset selection problem. The first one is a solution ranking method based on interval values to handle the noise in the objectives of this problem. The second one is a model estimation method for learning a joint probabilistic model of objectives and variables which is used to generate new solutions and advance through the search space. To simplify model estimation, l1 regularized regression is used to select a subset of problem variables before model learning. The proposed algorithm is compared with a well-known ranking method for interval-valued objectives and a standard multiobjective genetic algorithm. Particularly, the effects of the two new techniques are experimentally investigated. The experimental results show that the proposed algorithm is able to obtain comparable or better performance on the tested datasets.

Inteligencia artificial para la predicción y control del acabado superficial en procesos de fresado a alta velocidad

El objetivo principal de esta tesis doctoral es profundizar en el análisis y diseño de un sistema inteligente para la predicción y control del acabado superficial en un proceso de fresado a alta velocidad, basado fundamentalmente en clasificadores Bayesianos, con el prop´osito de desarrollar una metodolog´ıa que facilite el diseño de este tipo de sistemas. El sistema, cuyo propósito es posibilitar la predicción y control de la rugosidad superficial, se compone de un modelo aprendido a partir de datos experimentales con redes Bayesianas, que ayudar´a a comprender los procesos dinámicos involucrados en el mecanizado y las interacciones entre las variables relevantes. Dado que las redes neuronales artificiales son modelos ampliamente utilizados en procesos de corte de materiales, también se incluye un modelo para fresado usándolas, donde se introdujo la geometría y la dureza del material como variables novedosas hasta ahora no estudiadas en este contexto. Por lo tanto, una importante contribución en esta tesis son estos dos modelos para la predicción de la rugosidad superficial, que se comparan con respecto a diferentes aspectos: la influencia de las nuevas variables, los indicadores de evaluación del desempeño, interpretabilidad. Uno de los principales problemas en la modelización con clasificadores Bayesianos es la comprensión de las enormes tablas de probabilidad a posteriori producidas. Introducimos un m´etodo de explicación que genera un conjunto de reglas obtenidas de árboles de decisión. Estos árboles son inducidos a partir de un conjunto de datos simulados generados de las probabilidades a posteriori de la variable clase, calculadas con la red Bayesiana aprendida a partir de un conjunto de datos de entrenamiento. Por último, contribuimos en el campo multiobjetivo en el caso de que algunos de los objetivos no se puedan cuantificar en números reales, sino como funciones en intervalo de valores. Esto ocurre a menudo en aplicaciones de aprendizaje automático, especialmente las basadas en clasificación supervisada. En concreto, se extienden las ideas de dominancia y frontera de Pareto a esta situación. Su aplicación a los estudios de predicción de la rugosidad superficial en el caso de maximizar al mismo tiempo la sensibilidad y la especificidad del clasificador inducido de la red Bayesiana, y no solo maximizar la tasa de clasificación correcta. Los intervalos de estos dos objetivos provienen de un m´etodo de estimación honesta de ambos objetivos, como e.g. validación cruzada en k rodajas o bootstrap.---ABSTRACT---The main objective of this PhD Thesis is to go more deeply into the analysis and design of an intelligent system for surface roughness prediction and control in the end-milling machining process, based fundamentally on Bayesian network classifiers, with the aim of developing a methodology that makes easier the design of this type of systems. The system, whose purpose is to make possible the surface roughness prediction and control, consists of a model learnt from experimental data with the aid of Bayesian networks, that will help to understand the dynamic processes involved in the machining and the interactions among the relevant variables. Since artificial neural networks are models widely used in material cutting proceses, we include also an end-milling model using them, where the geometry and hardness of the piecework are introduced as novel variables not studied so far within this context. Thus, an important contribution in this thesis is these two models for surface roughness prediction, that are then compared with respecto to different aspects: influence of the new variables, performance evaluation metrics, interpretability. One of the main problems with Bayesian classifier-based modelling is the understanding of the enormous posterior probabilitiy tables produced. We introduce an explanation method that generates a set of rules obtained from decision trees. Such trees are induced from a simulated data set generated from the posterior probabilities of the class variable, calculated with the Bayesian network learned from a training data set. Finally, we contribute in the multi-objective field in the case that some of the objectives cannot be quantified as real numbers but as interval-valued functions. This often occurs in machine learning applications, especially those based on supervised classification. Specifically, the dominance and Pareto front ideas are extended to this setting. Its application to the surface roughness prediction studies the case of maximizing simultaneously the sensitivity and specificity of the induced Bayesian network classifier, rather than only maximizing the correct classification rate. Intervals in these two objectives come from a honest estimation method of both objectives, like e.g. k-fold cross-validation or bootstrap.

A recursive algorithm to compute a basis of a similarity

Se definen conceptos y se aplica el teorema de Valverde para escribir un algoritmo que computa bases de similaridades. This paper studies sorne theory and methods to build a representation theorem basis of a similarity from the basis of its subsimilarities, providing an alternative recursive method to compute the basis of a similarity.

Interval-based resource usage verification: Formalization and prototype

In an increasing number of applications (e.g., in embedded, real-time, or mobile systems) it is important or even essential to ensure conformance with respect to a specification expressing resource usages, such as execution time, memory, energy, or user-defined resources. In previous work we have presented a novel framework for data size-aware, static resource usage verification. Specifications can include both lower and upper bound resource usage functions. In order to statically check such specifications, both upper- and lower-bound resource usage functions (on input data sizes) approximating the actual resource usage of the program which are automatically inferred and compared against the specification. The outcome of the static checking of assertions can express intervals for the input data sizes such that a given specification can be proved for some intervals but disproved for others. After an overview of the approach in this paper we provide a number of novel contributions: we present a full formalization, and we report on and provide results from an implementation within the Ciao/CiaoPP framework (which provides a general, unified platform for static and run-time verification, as well as unit testing). We also generalize the checking of assertions to allow preconditions expressing intervals within which the input data size of a program is supposed to lie (i.e., intervals for which each assertion is applicable), and we extend the class of resource usage functions that can be checked.

A collaborative filtering similarity measure based on singularities.

Recommender systems play an important role in reducing the negative impact of informa- tion overload on those websites where users have the possibility of voting for their prefer- ences on items. The most normal technique for dealing with the recommendation mechanism is to use collaborative filtering, in which it is essential to discover the most similar users to whom you desire to make recommendations. The hypothesis of this paper is that the results obtained by applying traditional similarities measures can be improved by taking contextual information, drawn from the entire body of users, and using it to cal- culate the singularity which exists, for each item, in the votes cast by each pair of users that you wish to compare. As such, the greater the measure of singularity result between the votes cast by two given users, the greater the impact this will have on the similarity. The results, tested on the Movielens, Netflix and FilmAffinity databases, corroborate the excellent behaviour of the singularity measure proposed.

A Balanced Memory-Based Collaborative Filtering Similarity Measure.

Collaborative filtering recommender systems contribute to alleviating the problem of information overload that exists on the Internet as a result of the mass use of Web 2.0 applications. The use of an adequate similarity measure becomes a determining factor in the quality of the prediction and recommendation results of the recommender system, as well as in its performance. In this paper, we present a memory-based collaborative filtering similarity measure that provides extremely high-quality and balanced results; these results are complemented with a low processing time (high performance), similar to the one required to execute traditional similarity metrics. The experiments have been carried out on the MovieLens and Netflix databases, using a representative set of information retrieval quality measures.

SSONDE: Semantic Similarity On liNked Data Entities

Linked data offers a promising setting to encode, publish and share metadata of resources. As the matter of fact, it is already adopted by data producers such as European Environment Agency, US and some EU Governs, whose first ambition is to share (meta)data making their processes more effective and transparent. Such as an increasing interest and involvement of data providers surely represents a genuine witness of the web of data success, but in a longer perspective, frameworks supporting linked data consumers in their decision making processes will be a compelling need. In this respect, the talk is introducing SSONDE, a framework enabling in detailed comparison, ranking and selection of linked data resources through the analysis of their RDF ontology driven metadata. SSONDE implements an instance similarity especially designed to support in resource selection, namely the process stakeholders engage to choose a set of resources suitable for a given analysis purpose: (i) it deploys an asymmetric similarity assessment to emphasize information about gains and losses the stakeholders get adopting a resource in place of another; (ii) it relies on an explicit formalization of contexts to tailor the similarity assessment with respect to specific user-defined selection goals. The talk aims at providing an insight on SSONDE instance similarity and it will briefly describe some examples of SSONDE deployment in the context of linked data consumption.

A new similarity function for generalized trapezoidal fuzzy numbers

Numerous authors have proposed functions to quantify the degree of similarity between two fuzzy numbers using various descriptive parameters, such as the geometric distance, the distance between the centers of gravity or the perimeter. However, these similarity functions have drawback for specific situations. We propose a new similarity measure for generalized trapezoidal fuzzy numbers aimed at overcoming such drawbacks. This new measure accounts for the distance between the centers of gravity and the geometric distance but also incorporates a new term based on the shared area between the fuzzy numbers. The proposed measure is compared against other measures in the literature.

On the generative equations of fractal self-similarity in granular media and the related PSD models

The physical appearance of granular media suggests the existence of geometrical scale invariance. The paper discuss how this physico-empirical property can be mathematically encoded leading to different generative models: a smooth one encoded by a differential equation and another encoded by an equation coming from a measure theoretical property.

Species richness and similarity of vascular plants in the Spanish dehesas at two spatial scales

Aims of study: The goals of this paper are to summarize and to compare plant species richness and floristic similarity at two spatial scales; mesohabitat (normal, eutrophic, and oligotrophic dehesas) and dehesa habitat; and to establish guidelines for conserving species diversity in dehesas. Area of study: We considered four dehesa sites in the western Peninsular Spain, located along a climatic and biogeographic gradient from north to south. Main results: Average alpha richness for mesohabitats was 75.6 species, and average alpha richness for dehesa sites was 146.3. Gamma richness assessed for the overall dehesa habitat was 340.0 species. The species richness figures of normal dehesa mesohabitat were significantly lesser than of the eutrophic mesohabitat and lesser than the oligotrophic mesohabitat too. No significant differences were found for species richness among dehesa sites. We have found more dissimilarity at local scale (mesohabitat) than at regional scale (habitat). Finally, the results of the similarity assessment between dehesa sites reflected both climatic and biogeographic gradients. Research highlights: An effective conservation of dehesas must take into account local and regional conditions all along their distribution range for ensuring the conservation of the main vascular plant species assemblages as well as the associated fauna

Diseño de instrumentos para configurar una política industrial en España basada en los postulados de la economía evolutiva

La evidencia empírica aplicada a países de cierto tamaño y desarrollo económico, muestra que existe una relación directa y positiva entre la intensidad de la industria manufacturera, medida como porcentaje de su PIB, y ciertas variables económicas como, el crecimiento, el desempleo y la balanza exterior de bienes y servicios. En el caso de esta última, se verifica empíricamente, que los países con una proporción de actividad manufacturera inferior al 20%, tienen una marcada tendencia a presentar déficits crónicos de balanza de bienes y servicios, lo que conduce a persistentes déficits por cuenta corriente, al ser el primer déficit el principal componente del segundo. Esto trae consigo un continuado incremento del endeudamiento externo que no cesa, y que terminará en algún momento por desequilibrar el conjunto de la economía de los países con esos déficits crónicos. Las anteriores conclusiones, abren una vía de orientación de la política económica, que tiene como objetivo la promoción de la industria manufacturera de cada país. Y esto es un hecho ya en 2014. Países relevantes, como Alemania o Francia en la UE, incluso los EEUU y últimamente el Reino Unido, y por supuesto países del área asiática como Japón, Corea del Sur y China, llevan años promoviendo su industria manufacturera. Resulta significativo, que el debate ideológico sobre la bondad de la aplicación política industrial por parte de los gobiernos, frente a las teorías liberales de mantener a los poderes públicos lejos de ese tipo de actividades, haya dado paso a un modelo generalizado de corte más bien horizontal, donde los países casi sin excepciones apoyan el desarrollo de sus empresas con numerosos instrumentos, que van bastante más lejos de los habituales de I+D. Se valora por tanto, la industria manufacturera como algo vital para el equilibrio económico. Incluso la UE, defensora durante décadas de la no intervención de los diferentes Estados miembros en actividades de promoción industrial más allá del apoyo a las actividades de I+D, realiza un giro copernicano, que termina en 2012 proclamando que la industria manufacturera es vital para el equilibrio económico de la UE, que hay que promoverla, e incluso marca un objetivo, precisamente del 20%, como contribución manufacturera a su PIB. Es decir, se da por asumido que los servicios no son un sustituto indefinido de la industria y que por tanto tienen un límite, lo que se contrapone frontalmente contra la anterior creencia de que el aumento de la participación de los servicios en la economía, no solo era bueno, sino un síntoma de desarrollo. Esta premisa ya ha dejado de ser cierta para esos y otros países. En cambio, en España nada de esto sucede en las dos últimas décadas, sino que la industria manufacturera no recibe especial atención de los poderes públicos y se desliza en una pendiente de disminución de su contribución al PIB, que incluso se acelera con la crisis económica que comienza en 2007, hasta alcanzar cifras del orden del 12% del PIB en 2013. La política económica que se aplica es la de la deflación de costes, con los efectos consecuentes sobre los salarios y sobre la capacidad de la economía de generar riqueza. Se apuesta por un modelo de mano de obra barata, que recuerda al de los años 60. Como indicador relevante de esta situación, lo que exporta la industria manufacturera española, no ha ganado en contenido tecnológico en los últimos quince años. Esta situación se ve empeorada por un hecho significativo adicional, y es que casi el 40% de las ventas de la industria manufacturera española provienen de empresas de propiedad extranjera, con lo que eso supone por una parte de dependencia tecnológica del exterior como en el caso del automóvil, y de incertidumbre sobre su futuro, al estar basadas en el mantenimiento en el futuro de una mano de obra barata, que frenará que los españoles que trabajan en esas empresas, progresen económicamente. La propuesta de esta Tesis, es en cambio apostar por un modelo de crecimiento para España que tenga como uno de sus pilares el desarrollo de una industria manufacturera sólida y con cada vez mayor contenido tecnológico. Para ello, se propone un Plan de política industrial, donde se incluye la creación de actores impulsores de este plan, que deben ser del máximo nivel político. Si los diferentes gobiernos no entienden y asumen esta necesidad de apoyo a la industria a largo plazo e independiente de los cambios políticos, no será posible llevar a cabo este Plan. Para su puesta en marcha, se propone la creación o refuerzo de numerosos instrumentos de apoyo a la industria manufacturera de carácter fundamentalmente horizontal que van mucho más allá de los habituales del I+D, y que en varios casos, tienen una clara semejanza con otros existentes ya en otros países desarrollados desde hace años. La promoción de la industria manufacturera necesita nuevos instrumentos, como una financiación a largo plazo para las empresas, una promoción ordenada y eficaz de la actividad internacional de las empresas exportadoras, la mayoría de las cuales exportan productos manufacturados, una educación y formación profesional que esté alineada con estos objetivos, unos instrumentos que apoyen en especial el desarrollo la industria manufacturera, o la participación minoritaria pero significativa, del Estado en empresas españolas pertenecientes a sectores estratégicos entre otros. En resumen, esta Tesis propone una alternativa de política económica radicalmente diferente a la de dejar la industria manufacturera española a su suerte, y basar el futuro económico de España en una mano de obra barata. ABSTRACT The empirical evidence, applied to countries of certain size and economic development, shows that there exists a direct and positive relationship between industrial manufacturing activity, measured as a percentage of GDP, and certain economic variables, such as growth, unemployment and the foreign balance of trade. In the case of the latter, it is verified empirically that the countries with a percentage of manufacturing activity below 20% have a marked tendency for chronic deficits of the balance of trade, leading to persistent deficits in the current account, being that the former deficit is the main component of the latter. This brings about a continued increase in foreign debt that does not cease, and that will end at some point by disrupting the economy of the countries with these chronic deficits. The previous conclusions open the way to a new direction for economic policy, which promotes industrial manufacturing in each country. This is already a fact in 2014. Relevant countries, such as Germany or France in the EU, even the US and ultimately the UK, and of course countries of East Asia such as Japan, South Korea and China, have been promoting their industrial manufacturing for years. It becomes significant that the ideological debate about the goodwill of the application of industrial policy by governments, against liberal theories that maintain public powers far from these kinds of activities, has taken a step towards a horizontal-cut generalized model, where countries, with almost no exception, rely on various instruments to develop their companies that go much further than the usual R&D. Industrial manufacture is therefore valued as vital for economic stability. Even the EU, proponent for decades of non-intervention policy that goes beyond R&D, has gone full circle, ending in 2012 by proclaiming that industrial manufacture is vital for the economic stability of the EU, that it must be promoted. They even mark precisely 20% as an objective for manufacturing as a percentage of GDP. In other words, it is a given that services are not an indefinite substitute for industry, and that therefore it has a limit as such. This rejects the notion that the increase in services at the cost of manufacture is not only healthy, but is also a symptom of development. This premise is no longer true for these and other countries. On the other hand, none of this happens in Spain, where industrial manufacture receives no special attention from the public authorities, and it slides on a downward slope of percentage contribution to GDP, which accelerates the economic crisis that begins in 2007, until manufacture reaches values of around 12% of GDP in 2013. The economic policy applied is that of cost deflation, with consequential effects on wages and the capacity of the economy to generate wealth. A model is proposed for cheaper labor, akin to that of the 1960s. As a relevant indicator of this situation, manufacturing exports from Spain have not grown technologically in the last 15 years. The situation is made worse by another significant fact: almost 40% of sales of the manufacturing industry originate from companies of foreign origin, which supposes on one hand a technological dependence on foreign countries, such as in the case of the automotive industry, and on the other hand uncertainty in its future, being that they are based on maintaining cheap labor in the future, which will slow economic progress of Spaniards working in these companies. The proposition of this Thesis is to bet on a growth model for Spain that has as one of its pillars the development of a solid manufacturing industry, with increasing technological content. For this, an industrial policy plan is proposed, which includes the creation of driving agents for this plan, which must be of maximum political level. If the various governments don’t understand and assume this necessity for support of industry in the long term, independent of political change, this plan will not be accomplished. To start it, the creation or reinforcement of numerous instruments to promote the manufacturing activities are proposed, with a fundamentally horizontal nature that goes far beyond the usual R&D, and that, in several cases, have a clear similarity with others existing in other countries, having been developed for years. The promotion of the manufacturing industry needs new instruments, such as the long-term financing of companies, an orderly and efficient promotion of international activity of exporting companies, the most of which export manufactured goods, education and professional training which is in tune with these objectives, some instruments which support in particular the development of the manufacturing industry, or the minor yet significant participation of the State in Spanish companies belonging to strategic sectors, among others. In summary, this Thesis proposes an different alternative to the economic policy of leaving the manufacturing industry of Spain to its chances, and to base the economic future of Spain on a cheaper labor force.

Searching for a consensus similarity function for generalized trapezoidal fuzzy numbers

There is controversy regarding the use of the similarity functions proposed in the literature to compare generalized trapezoidal fuzzy numbers since conflicting similarity values are sometimes output for the same pair of fuzzy numbers. In this paper we propose a similarity function aimed at establishing a consensus. It accounts for the different approaches of all the similarity functions. It also has better properties and can easily incorporate new parameters for future improvements. The analysis is carried out on the basis of a large and representative set of pairs of trapezoidal fuzzy numbers.

Interval-based ranking in noisy evolutionary multiobjective optimization

As one of the most competitive approaches to multi-objective optimization, evolutionary algorithms have been shown to obtain very good results for many realworld multi-objective problems. One of the issues that can affect the performance of these algorithms is the uncertainty in the quality of the solutions which is usually represented with the noise in the objective values. Therefore, handling noisy objectives in evolutionary multi-objective optimization algorithms becomes very important and is gaining more attention in recent years. In this paper we present ?-degree Pareto dominance relation for ordering the solutions in multi-objective optimization when the values of the objective functions are given as intervals. Based on this dominance relation, we propose an adaptation of the non-dominated sorting algorithm for ranking the solutions. This ranking method is then used in a standardmulti-objective evolutionary algorithm and a recently proposed novel multi-objective estimation of distribution algorithm based on joint variable-objective probabilistic modeling, and applied to a set of multi-objective problems with different levels of independent noise. The experimental results show that the use of the proposed method for solution ranking allows to approximate Pareto sets which are considerably better than those obtained when using the dominance probability-based ranking method, which is one of the main methods for noise handling in multi-objective optimization.

Cardiometabolic responses to a battling rope high intensity interval training protocol

Trata de una conferencia invitada que ganó premio a la mejor comunicación científica.

Biometric authentication of users through iris by using key binding and similarity preserving hash functions = Autenticación biométrica de usuarios a través del iris mediante la ocultación de claves y funciones resumen que preservan la similitud

El extraordinario auge de las nuevas tecnologías de la información, el desarrollo de la Internet de las Cosas, el comercio electrónico, las redes sociales, la telefonía móvil y la computación y almacenamiento en la nube, han proporcionado grandes beneficios en todos los ámbitos de la sociedad. Junto a éstos, se presentan nuevos retos para la protección y privacidad de la información y su contenido, como la suplantación de personalidad y la pérdida de la confidencialidad e integridad de los documentos o las comunicaciones electrónicas. Este hecho puede verse agravado por la falta de una frontera clara que delimite el mundo personal del mundo laboral en cuanto al acceso de la información. En todos estos campos de la actividad personal y laboral, la Criptografía ha jugado un papel fundamental aportando las herramientas necesarias para garantizar la confidencialidad, integridad y disponibilidad tanto de la privacidad de los datos personales como de la información. Por otro lado, la Biometría ha propuesto y ofrecido diferentes técnicas con el fin de garantizar la autentificación de individuos a través del uso de determinadas características personales como las huellas dáctilares, el iris, la geometría de la mano, la voz, la forma de caminar, etc. Cada una de estas dos ciencias, Criptografía y Biometría, aportan soluciones a campos específicos de la protección de datos y autentificación de usuarios, que se verían enormemente potenciados si determinadas características de ambas ciencias se unieran con vistas a objetivos comunes. Por ello es imperativo intensificar la investigación en estos ámbitos combinando los algoritmos y primitivas matemáticas de la Criptografía con la Biometría para dar respuesta a la demanda creciente de nuevas soluciones más técnicas, seguras y fáciles de usar que potencien de modo simultáneo la protección de datos y la identificacíón de usuarios. En esta combinación el concepto de biometría cancelable ha supuesto una piedra angular en el proceso de autentificación e identificación de usuarios al proporcionar propiedades de revocación y cancelación a los ragos biométricos. La contribución de esta tesis se basa en el principal aspecto de la Biometría, es decir, la autentificación segura y eficiente de usuarios a través de sus rasgos biométricos, utilizando tres aproximaciones distintas: 1. Diseño de un esquema criptobiométrico borroso que implemente los principios de la biometría cancelable para identificar usuarios lidiando con los problemas acaecidos de la variabilidad intra e inter-usuarios. 2. Diseño de una nueva función hash que preserva la similitud (SPHF por sus siglas en inglés). Actualmente estas funciones se usan en el campo del análisis forense digital con el objetivo de buscar similitudes en el contenido de archivos distintos pero similares de modo que se pueda precisar hasta qué punto estos archivos pudieran ser considerados iguales. La función definida en este trabajo de investigación, además de mejorar los resultados de las principales funciones desarrolladas hasta el momento, intenta extender su uso a la comparación entre patrones de iris. 3. Desarrollando un nuevo mecanismo de comparación de patrones de iris que considera tales patrones como si fueran señales para compararlos posteriormente utilizando la transformada de Walsh-Hadarmard. Los resultados obtenidos son excelentes teniendo en cuenta los requerimientos de seguridad y privacidad mencionados anteriormente. Cada uno de los tres esquemas diseñados han sido implementados para poder realizar experimentos y probar su eficacia operativa en escenarios que simulan situaciones reales: El esquema criptobiométrico borroso y la función SPHF han sido implementados en lenguaje Java mientras que el proceso basado en la transformada de Walsh-Hadamard en Matlab. En los experimentos se ha utilizado una base de datos de imágenes de iris (CASIA) para simular una población de usuarios del sistema. En el caso particular de la función de SPHF, además se han realizado experimentos para comprobar su utilidad en el campo de análisis forense comparando archivos e imágenes con contenido similar y distinto. En este sentido, para cada uno de los esquemas se han calculado los ratios de falso negativo y falso positivo. ABSTRACT The extraordinary increase of new information technologies, the development of Internet of Things, the electronic commerce, the social networks, mobile or smart telephony and cloud computing and storage, have provided great benefits in all areas of society. Besides this fact, there are new challenges for the protection and privacy of information and its content, such as the loss of confidentiality and integrity of electronic documents and communications. This is exarcebated by the lack of a clear boundary between the personal world and the business world as their differences are becoming narrower. In both worlds, i.e the personal and the business one, Cryptography has played a key role by providing the necessary tools to ensure the confidentiality, integrity and availability both of the privacy of the personal data and information. On the other hand, Biometrics has offered and proposed different techniques with the aim to assure the authentication of individuals through their biometric traits, such as fingerprints, iris, hand geometry, voice, gait, etc. Each of these sciences, Cryptography and Biometrics, provides tools to specific problems of the data protection and user authentication, which would be widely strengthen if determined characteristics of both sciences would be combined in order to achieve common objectives. Therefore, it is imperative to intensify the research in this area by combining the basics mathematical algorithms and primitives of Cryptography with Biometrics to meet the growing demand for more secure and usability techniques which would improve the data protection and the user authentication. In this combination, the use of cancelable biometrics makes a cornerstone in the user authentication and identification process since it provides revocable or cancelation properties to the biometric traits. The contributions in this thesis involve the main aspect of Biometrics, i.e. the secure and efficient authentication of users through their biometric templates, considered from three different approaches. The first one is designing a fuzzy crypto-biometric scheme using the cancelable biometric principles to take advantage of the fuzziness of the biometric templates at the same time that it deals with the intra- and inter-user variability among users without compromising the biometric templates extracted from the legitimate users. The second one is designing a new Similarity Preserving Hash Function (SPHF), currently widely used in the Digital Forensics field to find similarities among different files to calculate their similarity level. The function designed in this research work, besides the fact of improving the results of the two main functions of this field currently in place, it tries to expand its use to the iris template comparison. Finally, the last approach of this thesis is developing a new mechanism of handling the iris templates, considering them as signals, to use the Walsh-Hadamard transform (complemented with three other algorithms) to compare them. The results obtained are excellent taking into account the security and privacy requirements mentioned previously. Every one of the three schemes designed have been implemented to test their operational efficacy in situations that simulate real scenarios: The fuzzy crypto-biometric scheme and the SPHF have been implemented in Java language, while the process based on the Walsh-Hadamard transform in Matlab. The experiments have been performed using a database of iris templates (CASIA-IrisV2) to simulate a user population. The case of the new SPHF designed is special since previous to be applied i to the Biometrics field, it has been also tested to determine its applicability in the Digital Forensic field comparing similar and dissimilar files and images. The ratios of efficiency and effectiveness regarding user authentication, i.e. False Non Match and False Match Rate, for the schemes designed have been calculated with different parameters and cases to analyse their behaviour.