[pt] Os Projetos de Eficiência Energética Gerenciados pelo Lado da Demanda (GLD), bem como todo projeto de engenharia, requerem decisões de investimentos que possuem incertezas associadas. As incertezas econômicas devem-se a fatores exógenos ao projeto sendo, em geral, representadas por oscilações estocásticas dos custos da energia elétrica. As incertezas técnicas estão associadas a fatores internos, como o desempenho dos projetos em função da tecnologia eficiente escolhida, da sua operação e manutenção. A decisão dos clientes e investidores na aquisição de Projetos de Eficiência Energética depende do retorno esperado nos ganhos com a energia economizada, como por exemplo, na venda desta energia no mercado de curto prazo. Esta tese investiga uma nova metodologia que, considerando as incertezas técnicas e econômicas, efetua uma análise mais abrangente e realista do cenário complexo de negócios que envolvem os Projetos de Eficiência Energética no Brasil. A metodologia contribui para a tomada de decisão considerando a flexibilidade gerencial e a avaliação dos riscos específicos dos projetos. Esta se baseia em técnicas inteligentes para a otimização de diagnósticos energéticos associados à análise de opções reais e avaliação econômica de Projetos de Eficiência Energética Gerenciados pelo Lado da Demanda (GLD), aplicados ao setor de energia elétrica no Brasil. A metodologia é avaliada em dois Projetos de Eficiência Energética, para os usos finais de Iluminação e Climatização de Ambientes, em uma unidade consumidora da classe Comercial, localizada na Cidade do Rio de Janeiro e conectada ao sistema de distribuição em Média Tensão (13,8kV). Os resultados revelaram que a partir da aplicação de Algoritmos Genéticos na otimização de diagnósticos energéticos puderam-se construir subprojetos originados de um projeto maior, mantendo-se, ou até ampliando-se a Relação Custo vs. Beneficio (RCB). E, desta forma, contribuir para a viabilização de alternativas ótimas de projetos que incentivam a aplicação da Eficiência Energética no Brasil. / [en] The Energy Efficiency Projects Managed by Demand Side (DSM), as well as all engineering design, require investment decisions that have associated uncertainties. Economic uncertainties are due to factors exogenous to the project being generally represented by stochastic fluctuations of electricity costs. The technical uncertainties are associated with internal factors such as performance of the projects on the basis of efficient technology chosen, its operation and maintenance. The decision of customers and investors in the acquisition of Energy Efficiency Projects depends on the expected return on the earnings of the energy saved, for example, the sale of this energy in the short term. This thesis investigates a new methodology which, considering the technical and economic uncertainties, performs a more comprehensive and realistic business complex scenario involving the Energy Efficiency Projects in Brazil. The methodology helps decision making considering managerial flexibility and risk assessment of specific projects. This is based on intelligent techniques for optimizing energy diagnoses associated with real options analysis and economic evaluation of Energy Efficiency Projects Managed by Demand Side (DSM), applied to the electricity sector in Brazil. The methodology is evaluated in two Energy Efficiency Projects for the end uses of lighting and Air Conditioning, in a consumer unit of the Commercial category, located in the city of Rio de Janeiro and connected to the distribution system in Medium Voltage (13.8kV). The results showed that with the application of genetic algorithms in optimization of energy diagnoses subprojects originated from a larger project could be built, maintaining or even widening the Cost vs. Value. Benefit (RCB) ratio. And in this way, contribute to the viability of alternative optimal designs that encourage the implementation of Energy Efficiency in Brazil.

No / Evolutionary membrane computing is an important research direction of membrane computing that aims to explore the complex interactions between membrane computing and evolutionary computation. These disciplines are receiving increasing attention. In this paper, an overview of the evolutionary membrane computing state-of-the-art and new results on two established topics in well defined scopes (membrane-inspired evolutionary algorithms and automated design of membrane computing models) are presented. We survey their theoretical developments and applications, sketch the differences between them, and compare the advantages and limitations. (C) 2014 Elsevier Inc. All rights reserved.

Computer-generated music composition programs have yet to produce creative, natural sounding music. To date, most approaches constrain the search space heuristically while ignoring the inherent structure of music over time. To address this problem, this thesis introduces NEAT Drummer, which evolves a special kind of artificial neural network (ANN) called compositional pattern producing networks (CPPNs) with the NeuroEvolution of Augmenting Topologies (NEAT) method for evolving increasingly complex structures. CPPNs in NEAT Drummer input existing human compositions and output an accompanying drum track. The existing musical parts form a scaffold i.e. support structure, for the drum pattern outputs, thereby exploiting the functional relationship of drums to musical parts (e.g. to lead guitar, bru:is, etc.) The results are convincing drum patterns that follow the contours of the original song, validating a new approach to computergenerated music composition.

Tesis por compendio / [ES] Los videojuegos son proyectos multidisciplinares que implican, en buena medida, el desarrollo de software. Esta tesis trata la faceta del desarrollo de videojuegos relativa al software mediante la Ingeniería del Software basada en Búsqueda (SBSE, Search-based Software Engineering). El objetivo específico de este trabajo es valerse de las características de los videojuegos en pro de una Ingeniería del Software de Videojuegos basada en Búsqueda (SBGSE, Search-based Game Software Engineering), incluyendo el uso de simulaciones de videojuegos para guiar búsquedas, codificación de granularidad fina y operaciones genéticas de mejora. Las aproximaciones propuestas superan a las de referencia en mantenimiento (trazabilidad de requisitos) y creación de contenido (generación de NPCs). El mantenimiento y la creación de contenido son, a menudo, tareas esenciales para garantizar la retención de usuarios por medio de actualizaciones o expansiones. Además, esta investigación aborda la necesidad de estudios de caso industriales. Esta tesis presenta un compendio que incluye tres artículos realizados durante el proceso de investigación y publicados en revistas académicas, con resultados que muestran que las aproximaciones de la Ingeniería del Software de Videojuegos basada en Búsqueda (SBGSE, Search-based Game Software Engineering) pueden mejorar la calidad de las soluciones generadas, así como reducir el tiempo necesario para producirlas. / [CA] Els videojocs són projectes multidisciplinaris que impliquen, en bona part, el desenvolupament de software. Aquesta tesi tracta la faceta del desenvolupament de videojocs relativa al software mitjançant l'Enginyeria del Software basada en Cerca (SBSE, Search-based Software Engineering). L'objectiu específic d'aquest treball és valdre's de les característiques dels videojocs en pro d'una Enginyeria del Software de Videojocs basada en Cerca (SBGSE, Search-based Game Software Engineering), incloent-hi l'ús de simulacions de videojocs per a guiar cerques, codificació de granularitat fina i operacions genètiques de millora. Les aproximacions proposades superen a les de referència en manteniment (traçabilitat de requisits) i creació de contingut (generació de NPCs). El manteniment i la creació de contingut són, sovint, tasques essencials per a garantir la retenció d'usuaris per mitjà d'actualitzacions o expansions. A més, aquesta investigació aborda la necessitat d'estudis de cas industrials. Aquesta tesi presenta un compendi que inclou tres articles realitzats durant el procés d'investigació i publicats en revistes acadèmiques, amb resultats que mostren que les aproximacions de l'Enginyeria del Software de Videojocs basada en Cerca (SBGSE, Search-based Game Software Engineering) poden millorar la qualitat de les solucions generades, així com reduir el temps necessari per a produir-les. / [EN] Video games are multidisciplinary projects which involve software development to a significant extent. This thesis tackles the software aspect of video game development through Search-based Engineering. Specifically, the objective of this work is to leverage the characteristics of video games towards Search-based Game Software Engineering, including the use of video game simulations to guide the search, a fine-grained encoding, and improvement genetic operations. The approaches proposed outperform the baselines in maintenance (requirement traceability) and content creation (NPC generation) tasks. Maintenance and content creation are often essential tasks to ensure player retention by means of updates or expansions. In addition, this research addresses the need for industrial case studies. This thesis presents a compendium that includes three papers produced through the research and published in academic journals, with results that show that Search-based Game Software Engineering approaches can provide improved solutions, in terms of quality and time cost. / This work has been partially supported by the Ministry of Economy and Competitiveness (MINECO) through the Spanish National R+D+i Plan and ERDF funds under the Project ALPS (RTI2018-096411-B-I00). / Blasco Latorre, D. (2024). Towards Search-based Game Software Engineering [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/203655 / Compendio

Η παρούσα διατριβή ασχολείται με τη μελέτη, την ανάπτυξη και τη θεμελίωση νέων μεθόδων Υπολογιστικής Νοημοσύνης και Ευφυούς Βελτιστοποίησης. Συνοπτικά οργανώνεται στα ακόλουθα τρία μέρη: Αρχικά παρουσιάζεται το πεδίο της Υπολογιστικής Νοημοσύνης και πραγματοποιείται μία σύντομη αναφορά στους τρεις κύριους κλάδους της, τον Εξελικτικό Υπολογισμό, τα Τεχνητά Νευρωνικά Δίκτυα και τα Ασαφή Συστήματα. Το επόμενο μέρος αφιερώνεται στην παρουσίαση νέων, καινοτόμων οικογενειών των αλγορίθμων Βελτιστοποίησης Σμήνους Σωματιδίων (ΒΣΣ) και των Διαφοροεξελικτικών Αλγόριθμων (ΔΕΑ), για την επίλυση αριθμητικών προβλημάτων βελτιστοποίησης χωρίς περιορισμούς, έχοντας είτε ένα, είτε πολλαπλούς ολικούς βελτιστοποιητές. Οι αλγόριθμοι ΒΣΣ και ΔΕΑ αποτελούν τις βασικές μεθοδολογίες της παρούσας διατριβής. Όλες οι οικογένειες μεθόδων που προτείνονται, βασίζονται σε παρατηρήσεις των κοινών δομικών χαρακτηριστικών των ΒΣΣ και ΔΕΑ, ενώ η κάθε προτεινόμενη οικογένεια τις αξιοποιεί με διαφορετικό τρόπο, δημιουργώντας νέες, αποδοτικές μεθόδους με αρκετά ενδιαφέρουσες ιδιότητες και δυναμική. Η παρουσίαση του ερευνητικού έργου της διατριβής ολοκληρώνεται με το τρίτο μέρος στο οποίο περιλαμβάνεται μελέτη και ανάπτυξη μεθόδων ολικής βελτιστοποίησης για την εκπαίδευση Τεχνητών Νευρωνικών Δικτύων Υψηλής Τάξης, σε σειριακά και παράλληλα ή / και κατανεμημένα υπολογιστικά συστήματα. Η διδακτορική διατριβή ολοκληρώνεται με βασικά συμπεράσματα και τη συνεισφορά της. / The main subject of the thesis at hand revolves mainly around the development and foundations of new methods of computational intelligence and intelligent optimization. The thesis is organized into the following three parts: Firstly, we briefly present an overview of the field of Computational Intelligence, by describing its main categories, the Evolutionary Computation, the Artificial Neural Networks and the Fuzzy Systems. In the second part, we provide a detailed description of the newly developed families of algorithms for solving unconstrained numerical optimization problems in continues spaces with at least one global optimum. The proposed families are based on two well-known and widely used algorithms, namely the Particle Swarm Optimization (PSO) and the Differential Evolution (DE) algorithm. Both DE and PSO are the basic components for almost all methodologies proposed in the thesis. The proposed methodologies are based on common observations of the dynamics, the structural and the spacial characteristics of DE and PSO algorithms. Four novel families are presented in this part which exploit the aforementioned characteristics of the DE and the PSO algorithms. The proposed methodologies are efficient methods with quite interesting properties and dynamics. The presentation and description of our research contribution ends with the third and last part of the thesis, which includes the study and the development of novel global optimization methodologies for training Higher order Artificial Neural Networks in serial and parallel / distributed computational environments. The thesis ends with a brief summary, conclusions and discussion of the contribution of this thesis.

[pt] A programação de produção em refinaria (scheduling) pode ser compreendida como uma sequência de decisões que buscam otimizar a alocação de recursos, o sequenciamento de atividades e a realização temporal dessas atividades, respeitando um conjunto de restrições de diferentes naturezas e visando o atendimento de múltiplos objetivos onde fatores como atendimento à demanda de produção e minimização de variações operacionais nos equipamentos coexistem na mesma função. Este trabalho propõe o uso da técnica de Programação Genética para automatizar a criação de programas que representem uma solução completa de programação de petróleo em uma refinaria dentro de um horizonte de tempo. Para a evolução destes programas foi desenvolvida uma linguagem específica para o domínio de problemas de scheduling de petróleo e aplicada de forma a representar as principais atividades do estudo de caso. Para tal, a primeira etapa consistiu da avaliação de alguns cenários de programação de produção de forma a selecionar as atividades que devessem ser representadas e como fazê-lo. No modelo proposto, o cromossomo quântico guarda a superposição de estados de todas as soluções possíveis e, através do processo evolutivo e observação dos genes quânticos, o cromossomo clássico é criado como uma sequencia linear de instruções a serem executadas. As instruções executadas representam o scheduling. A orientação deste processo é feita através de uma função de aptidão multiobjetivo que hierarquiza as avaliações sobre o tempo de operação das unidades de destilação, o prazo para descarregamento de navios, a utilização do duto que movimenta óleo entre terminal e refinaria, além de fatores como número de trocas de tanques e uso de tanques de injeção nas unidades de destilação. No desenvolvimento deste trabalho foi contemplado um estudo sobre o conjunto de parâmetros para o modelo desenvolvido com base em um dos cenários de programação selecionados. A partir desta definição, para avaliação do modelo proposto, foram executadas diversas rodadas para cinco cenários de programação de petróleo. Os resultados obtidos foram comparados com estudo desenvolvido usando algoritmos genéticos cujas atividades, no cromossomo, possuem representação por ordem. A programação genética apresentou percentual de soluções aceitas variando entre 25 por cento e 90 por cento dependendo da complexidade do cenário, sendo estes valores superiores ao obtido usando Algoritmos Genéticos em todos os cenários, com esforço computacional menor. / [en] Refinery scheduling can be understood as a sequence of decisions that targets the optimization of available resources, sequencing and execution of activities on proper timing; always respecting restrictions of different natures. The final result must achieve multiple objectives guaranteeing co-existence of different factors in the same function, such as production demand fullfillment and minimize operational variation. In this work it is proposed the use of the genetic programming technique to automate the building process of programs that represent a complete oil scheduling solution within a defined time horizon. For the evolution of those programs, it was developed a domain specific language to translate oil scheduling instructions that was applied to represent the most relevant activities for the proposed case studies. For that, purpose first step was to evaluate a few real scheduling scenarios to select which activities needed to be represented and how to do that. On the proposed model, each quantum chromosome represents the overlapping of all solutions and by the evolutionary process (and quantum gene measurement) the classic chromosome is created as a linear sequence of scheduling instructions to be executed. The orientation for this process is performed through a multi-object fitness function that prioritizes the evaluations according to: the operating time of the atmospheric distillation unities, the oil unloading time from the ships, the oil pipeline operation to transport oil to the refinery and other parameters like the number of charge tanks switchover and injection tank used for the distillation unities. The scope of this work also includes a study about tuning for the developed model based in one of the considered scenarios. From this set, an evaluation of other different scheduling scenarios was performed to test the model. The obtained results were then compared with a developed model that uses genetic algorithms with order representation for the activities. The proposed model showed between 25 percent - 90 percent of good solutions depending on the scenario complexity. Those results exhibit higher percentage of good solutions requiring less computational effort than the ones obtained with the genetic algorithms.

Data mining (DM) is the process of finding patterns and relationships in databases. The breakthrough in computer technologies triggered a massive growth in data collected and maintained by organisations. In many applications, these data arrive continuously in large volumes as a sequence of instances known as a data stream. Mining these data is known as stream data mining. Due to the large amount of data arriving in a data stream, each record is normally expected to be processed only once. Moreover, this process can be carried out on different sites in the organisation simultaneously making the problem distributed in nature. Distributed stream data mining poses many challenges to the data mining community including scalability and coping with changes in the underlying concept over time. In this thesis, the author hypothesizes that learning classifier systems (LCSs) - a class of classification algorithms - have the potential to work efficiently in distributed stream data mining. LCSs are an incremental learner, and being evolutionary based they are inherently adaptive. However, they suffer from two main drawbacks that hinder their use as fast data mining algorithms. First, they require a large population size, which slows down the processing of arriving instances. Second, they require a large number of parameter settings, some of them are very sensitive to the nature of the learning problem. As a result, it becomes difficult to choose a right setup for totally unknown problems. The aim of this thesis is to attack these two problems in LCS, with a specific focus on UCS - a supervised evolutionary learning classifier system. UCS is chosen as it has been tested extensively on classification tasks and it is the supervised version of XCS, a state of the art LCS. In this thesis, the architectural design for a distributed stream data mining system will be first introduced. The problems that UCS should face in a distributed data stream task are confirmed through a large number of experiments with UCS and the proposed architectural design. To overcome the problem of large population sizes, the idea of using a Neural Network to represent the action in UCS is proposed. This new system - called NLCS { was validated experimentally using a small fixed population size and has shown a large reduction in the population size needed to learn the underlying concept in the data. An adaptive version of NLCS called ANCS is then introduced. The adaptive version dynamically controls the population size of NLCS. A comprehensive analysis of the behaviour of ANCS revealed interesting patterns in the behaviour of the parameters, which motivated an ensemble version of the algorithm with 9 nodes, each using a different parameter setting. In total they cover all patterns of behaviour noticed in the system. A voting gate is used for the ensemble. The resultant ensemble does not require any parameter setting, and showed better performance on all datasets tested. The thesis concludes with testing the ANCS system in the architectural design for distributed environments proposed earlier. The contributions of the thesis are: (1) reducing the UCS population size by an order of magnitude using a neural representation; (2) introducing a mechanism for adapting the population size; (3) proposing an ensemble method that does not require parameter setting; and primarily (4) showing that the proposed LCS can work efficiently for distributed stream data mining tasks.

Data mining (DM) is the process of finding patterns and relationships in databases. The breakthrough in computer technologies triggered a massive growth in data collected and maintained by organisations. In many applications, these data arrive continuously in large volumes as a sequence of instances known as a data stream. Mining these data is known as stream data mining. Due to the large amount of data arriving in a data stream, each record is normally expected to be processed only once. Moreover, this process can be carried out on different sites in the organisation simultaneously making the problem distributed in nature. Distributed stream data mining poses many challenges to the data mining community including scalability and coping with changes in the underlying concept over time. In this thesis, the author hypothesizes that learning classifier systems (LCSs) - a class of classification algorithms - have the potential to work efficiently in distributed stream data mining. LCSs are an incremental learner, and being evolutionary based they are inherently adaptive. However, they suffer from two main drawbacks that hinder their use as fast data mining algorithms. First, they require a large population size, which slows down the processing of arriving instances. Second, they require a large number of parameter settings, some of them are very sensitive to the nature of the learning problem. As a result, it becomes difficult to choose a right setup for totally unknown problems. The aim of this thesis is to attack these two problems in LCS, with a specific focus on UCS - a supervised evolutionary learning classifier system. UCS is chosen as it has been tested extensively on classification tasks and it is the supervised version of XCS, a state of the art LCS. In this thesis, the architectural design for a distributed stream data mining system will be first introduced. The problems that UCS should face in a distributed data stream task are confirmed through a large number of experiments with UCS and the proposed architectural design. To overcome the problem of large population sizes, the idea of using a Neural Network to represent the action in UCS is proposed. This new system - called NLCS { was validated experimentally using a small fixed population size and has shown a large reduction in the population size needed to learn the underlying concept in the data. An adaptive version of NLCS called ANCS is then introduced. The adaptive version dynamically controls the population size of NLCS. A comprehensive analysis of the behaviour of ANCS revealed interesting patterns in the behaviour of the parameters, which motivated an ensemble version of the algorithm with 9 nodes, each using a different parameter setting. In total they cover all patterns of behaviour noticed in the system. A voting gate is used for the ensemble. The resultant ensemble does not require any parameter setting, and showed better performance on all datasets tested. The thesis concludes with testing the ANCS system in the architectural design for distributed environments proposed earlier. The contributions of the thesis are: (1) reducing the UCS population size by an order of magnitude using a neural representation; (2) introducing a mechanism for adapting the population size; (3) proposing an ensemble method that does not require parameter setting; and primarily (4) showing that the proposed LCS can work efficiently for distributed stream data mining tasks.

Data mining (DM) is the process of finding patterns and relationships in databases. The breakthrough in computer technologies triggered a massive growth in data collected and maintained by organisations. In many applications, these data arrive continuously in large volumes as a sequence of instances known as a data stream. Mining these data is known as stream data mining. Due to the large amount of data arriving in a data stream, each record is normally expected to be processed only once. Moreover, this process can be carried out on different sites in the organisation simultaneously making the problem distributed in nature. Distributed stream data mining poses many challenges to the data mining community including scalability and coping with changes in the underlying concept over time. In this thesis, the author hypothesizes that learning classifier systems (LCSs) - a class of classification algorithms - have the potential to work efficiently in distributed stream data mining. LCSs are an incremental learner, and being evolutionary based they are inherently adaptive. However, they suffer from two main drawbacks that hinder their use as fast data mining algorithms. First, they require a large population size, which slows down the processing of arriving instances. Second, they require a large number of parameter settings, some of them are very sensitive to the nature of the learning problem. As a result, it becomes difficult to choose a right setup for totally unknown problems. The aim of this thesis is to attack these two problems in LCS, with a specific focus on UCS - a supervised evolutionary learning classifier system. UCS is chosen as it has been tested extensively on classification tasks and it is the supervised version of XCS, a state of the art LCS. In this thesis, the architectural design for a distributed stream data mining system will be first introduced. The problems that UCS should face in a distributed data stream task are confirmed through a large number of experiments with UCS and the proposed architectural design. To overcome the problem of large population sizes, the idea of using a Neural Network to represent the action in UCS is proposed. This new system - called NLCS { was validated experimentally using a small fixed population size and has shown a large reduction in the population size needed to learn the underlying concept in the data. An adaptive version of NLCS called ANCS is then introduced. The adaptive version dynamically controls the population size of NLCS. A comprehensive analysis of the behaviour of ANCS revealed interesting patterns in the behaviour of the parameters, which motivated an ensemble version of the algorithm with 9 nodes, each using a different parameter setting. In total they cover all patterns of behaviour noticed in the system. A voting gate is used for the ensemble. The resultant ensemble does not require any parameter setting, and showed better performance on all datasets tested. The thesis concludes with testing the ANCS system in the architectural design for distributed environments proposed earlier. The contributions of the thesis are: (1) reducing the UCS population size by an order of magnitude using a neural representation; (2) introducing a mechanism for adapting the population size; (3) proposing an ensemble method that does not require parameter setting; and primarily (4) showing that the proposed LCS can work efficiently for distributed stream data mining tasks.

Els sistemes de conversió de text en parla (CTP-SU) s'encarreguen de produir veu sintètica a partir d'un text d'entrada. Els CTP basats en selecció d'unitats (CTP-SU) recuperen la millor seqüència d'unitats de veu enregistrades prèviament en una base de dades (corpus). La recuperació es realitza mitjançant algorismes de programació dinàmica i una funció de cost ponderada. La ponderació de la funció de cost es realitza típicament de forma manual per part d'un expert. No obstant, l'ajust manual resulta costós des d'un punt de vista de coneixement prèvi, i imprecís en la seva execució. Per tal d'ajustar els pesos de la funció de cost, aquesta tesi parteix de la prova de viabilitat d'ajust perceptiu presentada per Alías (2006) que empra algorismes genètics interactius actius (active interactive Genetic Algorithm - aiGA). Aquesta tesi doctoral investiga les diferents problemàtiques que es presenten en aplicar els aiGAs en l'ajust de pesos d'un CTP-SU en un context real de selecció d'unitats. Primerament la tesi realitza un estudi de l'estat de l'art en l'ajust de pesos. Tot seguit, repassa la idoneïtat de la computació evolutiva interactiva per realitzar l'ajust revisant amb profunditat el treball previ. Llavors es presenten i es validen les propostes de millora. Les quatre línies mestres que guien les contribucions d'aquesta tesi són: la precisió en l'ajust dels pesos, la robustesa dels pesos obtinguts, l'aplicabilitat de la metodologia per qualsevol funció de cost i el consens dels pesos obtinguts incorporant el criteri de diferents usuaris. En termes de precisió la tesi proposa realitzar l'ajust perceptiu per diferents tipus (clústers) d'unitats respectant les seves peculiaritats fonètiques i contextuals. En termes de robustesa la tesi incorpora diferents mètriques evolutives (indicadors) que avaluen aspectes com l'ambigüitat en la cerca, la convergència d'un usuari o el nivell de consens entre diferents usuaris. Posteriorment, per estudiar l'aplicabilitat de la metodologia proposada s'ajusten perceptivament diferents pesos que combinen informació lingüística i simbòlica. La última contribució d'aquesta tesi estudia l'idoneïtat dels models latents per modelar les preferències dels diferents usuaris i obtenir una solució de consens. Paral•lelament, per fer el pas d'una prova de viabilitat a un entorn real de selecció d'unitats es treballa amb un corpus d'extensió mitjana (1.9h) etiquetat automàticament. La tesi permet concloure que l'aiGA a nivell de clúster és una metodologia altament competitiva respecte les altres tècniques d'ajust presents en l'estat de l'art. / Los sistemas de conversión texto-habla (CTH-SU) se encargan de producir voz sintética a partir de un texto de entrada. Los CTH basados en selección de unidades (CTH-SU) recuperan la mejor secuencia de unidades de voz grabadas previamente en una base de datos (corpus). La recuperación se realitza mediante algoritmos de programación dinámica y una función de coste ponderada. La ponderación de la función de coste se realiza típicamente de forma manual por parte de un experto. Sin embargo, el ajuste manual resulta costoso desde un punto de vista de conocimiento previo e impreciso en su ejecución. Para ajustar los pesos de la función de coste, esta tesis parte de la prueba de viabilidad de ajuste perceptivo presentada por Alías (2006) que emplea algoritmos genéticos interactivos activos (active interactive Genetic Algorithm - aiGA). Esta tesis doctoral investiga las diferentes problemáticas que se presentan al aplicar los aiGAs en el ajuste de pesos de un CTH-SU en un contexto real de selección de unidades. Primeramente la tesis realiza un estudio del estado del arte en el ajuste de pesos, posteriormente repasa la idoneidad de la computación evolutiva interactiva para realizar el ajuste revisando en profundidad el trabajo previo. Entonces se presentan y se validan las propuestas de mejora. Las cuatro líneas maestras que guían las contribuciones de esta tesis son: la precisión en el ajuste de los pesos, la robustez de los pesos obtenidos, la aplicabilidad de la metodología para cualquier función de coste y el consenso de los pesos obtenidos incorporando el criterio de diferentes usuarios. En términos de precisión la tesis propone realizar el ajuste perceptivo por diferentes tipos (clusters) de unidades respetando sus peculiaridades fonéticas y contextuales. En términos de robustez la tesis incorpora diferentes métricas evolutivas (indicadores) que evalúan aspectos como la ambigüedad en la búsqueda, la convergencia de un usuario o el nivel de consenso entre diferentes usuarios. Posteriormente, para estudiar la aplicabilidad de la metodología propuesta se ajustan perceptivamente diferentes pesos que combinan información lingüística y simbólica. La última contribución de esta tesis estudia la idoneidad de los modelos latentes para modelar las preferencias de los diferentes usuarios y obtener una solución de consenso. Paralelamente, para dar el paso de una prueba de viabilidad a un entorno real de selección de unidades se trabaja con un corpus de extensión media (1.9h) etiquetado automáticamente. La tesis permite concluir que el aiGA a nivel de cluster es una metodología altamente competitiva respecto a las otras técnicas de ajuste presentes en el estado del arte. / Text-to-Speech Systems (TTS) produce synthetic speech from an input text. Unit Selection TTS (US-TTS) systems are based on the retrieval of the best sequence of recorded speech units previously recorded into a database (corpus). The retrieval is done by means of dynamic programming algorithm and a weighted cost function. An expert typically performs the weighting of the cost function by hand. However, hand tuning is costly from a standpoint of previous training and inaccurate in terms of methodology. In order to properly tune the weights of the cost function, this thesis continues the perceptual tuning proposal submitted by Alías(2006) which uses active interactive Genetic Algorithms (aiGAs). This thesis conducts an investigation to the various problems that arise in applying aiGAs to the weight tuning of the cost function. Firstly, the thesis makes a deep revision to the state-of-the-art in weight tuning. Afterwards, the thesis outlines the suitability of Interactive Evolutionary Computation (IEC) to perform the weight tuning making a thorough review of previous work. Then, the proposals of improvement are presented. The four major guidelines pursued by this thesis are: accuracy in adjusting the weights, robustness of the weights obtained, the applicability of the methodology to any subcost distance and the consensus of weights obtained by different users. In terms of precision cluster-level perceptual tuning is proposed in order to obtain weights for different types (clusters) of units considering their phonetic and contextual properties. In terms of robustness of the evolutionary process, the thesis presents different metrics (indicators) to assess aspects such as the ambiguity within the evolutionary search, the convergence of one user or the level of consensus among different users. Subsequently, to study the applicability of the proposed methodology different weights are perceptually tuned combining linguistic and symbolic information. The last contribution of this thesis examines the suitability of latent models for modeling the preferences of different users and obtains a consensus solution. In addition, the experimentation is carried out through a medium size corpus (1.9h) automatically labelled in order fill the gap between the proof-of-principle and a real unit selection scenario. The thesis concludes that aiGAs are highly competitive in comparison to other weight tuning techniques from the state-of-the-art.