La prévision des séries temporelles est un problème qui est traité depuis de nombreuses années. On y trouve des applications dans différents domaines tels que : la finance, la médecine, le transport, etc. Dans cette thèse, on s’est intéressé aux méthodes issues de l’apprentissage artificiel : les réseaux de neurones et les SVM. On s’est également intéressé à l’intérêt des méta-méthodes pour améliorer les performances des prédicteurs, notamment l’approche locale. Dans une optique de diviser pour régner, les approches locales effectuent le clustering des données avant d’affecter les prédicteurs aux sous ensembles obtenus. Nous présentons une modification dans l’algorithme d’apprentissage des réseaux de neurones récurrents afin de les adapter à cette approche. Nous proposons également deux nouvelles techniques de clustering, la première basée sur les cartes de Kohonen et la seconde sur les arbres binaires. / Time series forecasting is a widely discussed issue for many years. Researchers from various disciplines have addressed it in several application areas : finance, medical, transportation, etc. In this thesis, we focused on machine learning methods : neural networks and SVM. We have also been interested in the meta-methods to push up the predictor performances, and more specifically the local models. In a divide and conquer strategy, the local models perform a clustering over the data sets before different predictors are affected into each obtained subset. We present in this thesis a new algorithm for recurrent neural networks to use them as local predictors. We also propose two novel clustering techniques suitable for local models. The first is based on Kohonen maps, and the second is based on binary trees.

[pt] A tarefa de Processamento de Linguagem Natural consiste em analisar linguagens naturais de forma computacional, facilitando o desenvolvimento de programas capazes de utilizar dados falados ou escritos. Uma das tarefas mais importantes deste campo é a Análise de Dependência. Tal tarefa consiste em analisar a estrutura gramatical de frases visando extrair aprender dados sobre suas relações de dependência. Em uma sentença, essas relações se apresentam em formato de árvore, onde todas as palavras são interdependentes. Devido ao seu uso em uma grande variedade de aplicações como Tradução Automática e Identificação de Papéis Semânticos, diversas pesquisas com diferentes abordagens são feitas nessa área visando melhorar a acurácia das árvores previstas. Uma das abordagens em questão consiste em encarar o problema como uma tarefa de classificação de tokens e dividi-la em três classificadores diferentes, um para cada sub-tarefa, e depois juntar seus resultados de forma incremental. As sub-tarefas consistem em classificar, para cada par de palavras que possuam relação paidependente, a classe gramatical do pai, a posição relativa entre os dois e a distância relativa entre as palavras. Porém, observando pesquisas anteriores nessa abordagem, notamos que o gargalo está na terceira sub-tarefa, a predição da distância entre os tokens. Redes Neurais Recorrentes são modelos que nos permitem trabalhar utilizando sequências de vetores, tornando viáveis problemas de classificação onde tanto a entrada quanto a saída do problema são sequenciais, fazendo delas uma escolha natural para o problema. Esse trabalho utiliza-se de Redes Neurais Recorrentes, em específico Long Short-Term Memory, para realizar a tarefa de predição da distância entre palavras que possuam relações de dependência como um problema de classificação sequence-to-sequence. Para sua avaliação empírica, este trabalho segue a linha de pesquisas anteriores e utiliza os dados do corpus em português disponibilizado pela Conference on Computational Natural Language Learning 2006 Shared Task. O modelo resultante alcança 95.27 por cento de precisão, resultado que é melhor do que o obtido por pesquisas feitas anteriormente para o modelo incremental. / [en] The Natural Language Processing task consists of analyzing the grammatical structure of a sentence written in natural language aiming to learn, identify and extract information related to its dependency structure. This data can be structured like a tree, since every word in a sentence has a head-dependent relation to another word from the same sentence. Since Dependency Parsing is used in many applications like Machine Translation, Semantic Role Labeling and Part-Of-Speech Tagging, researchers aiming to improve the accuracy on their models are approaching this task in many different ways. One of the approaches consists in looking at this task as a token classification problem, using different classifiers for each sub-task and joining them in an incremental way. These sub-tasks consist in classifying, for each head-dependent pair, the Part-Of-Speech tag of the head, the relative position between the two words and the distance between them. However, previous researches using this approach show that the bottleneck lies in the distance classifier. Recurrent Neural Networks are a kind of Neural Network that allows us to work using sequences of vectors, allowing for classification problems where both our input and output are sequences, making them a great choice for the problem at hand. This work studies the use of Recurrent Neural Networks, in specific Long Short-Term Memory networks, for the head-dependent distance classifier sub-task as a sequence-to-sequence classification problem. To evaluate its efficiency, this work follows the line of previous researches and makes use of the Portuguese corpus of the Conference on Computational Natural Language Learning 2006 Shared Task. The resulting model attains 95.27 percent precision, which is better than the previous results obtained using incremental models.

Language is the principal medium for ideas, while dialogue is the most natural and effective way for humans to interact with and access information from machines. Natural language generation (NLG) is a critical component of spoken dialogue and it has a significant impact on usability and perceived quality. Many commonly used NLG systems employ rules and heuristics, which tend to generate inflexible and stylised responses without the natural variation of human language. However, the frequent repetition of identical output forms can quickly make dialogue become tedious for most real-world users. Additionally, these rules and heuristics are not scalable and hence not trivially extensible to other domains or languages. A statistical approach to language generation can learn language decisions directly from data without relying on hand-coded rules or heuristics, which brings scalability and flexibility to NLG. Statistical models also provide an opportunity to learn in-domain human colloquialisms and cross-domain model adaptations. A robust and quasi-supervised NLG model is proposed in this thesis. The model leverages a Recurrent Neural Network (RNN)-based surface realiser and a gating mechanism applied to input semantics. The model is motivated by the Long-Short Term Memory (LSTM) network. The RNN-based surface realiser and gating mechanism use a neural network to learn end-to-end language generation decisions from input dialogue act and sentence pairs; it also integrates sentence planning and surface realisation into a single optimisation problem. The single optimisation not only bypasses the costly intermediate linguistic annotations but also generates more natural and human-like responses. Furthermore, a domain adaptation study shows that the proposed model can be readily adapted and extended to new dialogue domains via a proposed recipe. Continuing the success of end-to-end learning, the second part of the thesis speculates on building an end-to-end dialogue system by framing it as a conditional generation problem. The proposed model encapsulates a belief tracker with a minimal state representation and a generator that takes the dialogue context to produce responses. These features suggest comprehension and fast learning. The proposed model is capable of understanding requests and accomplishing tasks after training on only a few hundred human-human dialogues. A complementary Wizard-of-Oz data collection method is also introduced to facilitate the collection of human-human conversations from online workers. The results demonstrate that the proposed model can talk to human judges naturally, without any difficulty, for a sample application domain. In addition, the results also suggest that the introduction of a stochastic latent variable can help the system model intrinsic variation in communicative intention much better.

Made available in DSpace on 2016-08-17T14:53:01Z (GMT). No. of bitstreams: 1 Fabio Nogueira da Silva.pdf: 1098466 bytes, checksum: a72dcced91748fe6c54f3cab86c19849 (MD5) Previous issue date: 2008-06-20 / FUNDAÇÃO DE AMPARO À PESQUISA E AO DESENVOLVIMENTO CIENTIFICO E TECNOLÓGICO DO MARANHÃO / We present in this work the results about two neural networks methods to solve the algebraic Riccati(ARE), what are used in many applications, mainly in the Linear Quadratic Regulator (LQR), H2 and H1 controls. First is showed the real symmetric form of the ARE and two methods based on neural computation. One feedforward neural network (FNN), that de¯nes an error as function of the ARE and a recurrent neural network (RNN), which converts a constrain optimization problem, restricted to the state space model, into an unconstrained convex optimization problem de¯ning an energy as function of the ARE and Cholesky factor. A proposal to chose the learning parameters of the RNN used to solve the ARE, by making a surface of the parameters variations, thus we can tune the neural network for a better performance. Computational experiments related with the plant matrices perturbations of the tested systems in order to perform an analysis of the behavior of the presented methodologies, that are based on homotopies methods, where we chose a good initial condition and compare the results to the Schur method. Two 6th order systems were used, a Doubly Fed Induction Generator(DFIG) and an aircraft plant. The results showed the RNN a good alternative compared with the FNN and Schur methods. / Apresenta-se nesta dissertação os resultados a respeito de dois métodos neuronais para a resolução da equação algébrica de Riccati(EAR), que tem varias aplicações, sendo principalmente usada pelos Regulador Linear Quadrático(LQR), controle H2 e controle H1. É apresentado a EAR real e simétrica e dois métodos baseados em uma rede neuronal direta (RND) que tem a função de erro associada a EAR e uma rede neuronal recorrente (RNR) que converte um problema de otimização restrita ao modelo de espaço de estados em outro de otimização convexa em função da EAR e do fator de Cholesky de modo a usufruir das propriedades de convexidade e condições de otimalidade. Uma proposta para a escolha dos parâmetros da RNR usada para solucionar a EAR por meio da geração de superfícies com a variação paramétrica da RNR, podendo assim melhor sintonizar a rede neuronal para um melhor desempenho. Experimentos computacionais relacionados a perturbações nos sistemas foram realizados para analisar o comportamento das metodologias apresentadas, tendo como base o princípio dos métodos homotópicos, com uma boa condição inicial, a partir de uma ponto de operação estável e comparamos os resultados com o método de Schur. Foram usadas as plantas de dois sistemas: uma representando a dinâmica de uma aeronave e outra de um motor de indução eólico duplamente alimentado(DFIG), ambos sistemas de 6a ordem. Os resultados mostram que a RNR é uma boa alternativa se comparado com a RND e com o método de Schur.

A teoria de sistemas semidinâmicos impulsivos é um capítulo importante e moderno da teoria de sistemas dinâmicos topológicos. Sistemas impulsivos descrevem processos de evolução que sofrem variações de estado de curta duração e que podem ser consideradas instantâneas. Os sistemas impulsivos admitem vários fenômenos interessantes às vezes, por causa da sua irregularidade, e às vezes por causa da sua regularidade. Para muitos fenômenos naturais, os modelos determinísticos mais realistas são frequentemente descritos por sistemas que envolvem impulsos. Esta teoria vem sendo desenvolvida continuamente. O presente trabalho apresenta resultados originais sobre a teoria de conjuntos minimais, movimentos recorrentes, movimentos quase periódicos e fracamente quase periódicos, teoria de estabilidade de Lyapunov, teoria da quase estabilidade de Zhukovskij e, finalmente, a construção de trajetórias negativas para sistemas semidinâmicos com impulsos. Os resultados novos apresentados neste trabalho estão contidos em três artigos, dos quais dois já foram aceitos para publicação. Veja [13], [14] e [15] / The theory of impulsive semidynamical systems is an important and modern chapter of the theory of topological dynamical systems. Impulsive systems describe the evolution of process whose continuous dynamics are interrupted by abrupt changes of state. This kind of systems admits various interesting phenomena sometimes, because of their irregularity, and sometimes because of their regularity. In many natural phenomena, the real deterministic models are often described by systems which involve impulses. This theory has been developed continuously. This work presents original results involving the theory of minimal sets, recurrent motions, almost periodic and weakly almost periodic motions, the study of Lyapunov stability and Zhukovshij Quasi stability and the construction of negative trajectories for impulsive semidynamical systems. The new results presented in this work are contained in three papers namely [13], [14] and [15]

For the last twenty years, several assumptions have been expressed in the fields of information processing, neurophysiology and cognitive sciences. First, neural networks and their dynamical behaviors in terms of attractors is the natural way adopted by the brain to encode information. Any information item to be stored in the neural network should be coded in some way or another in one of the dynamical attractors of the brain, and retrieved by stimulating the network to trap its dynamics in the desired item’s basin of attraction. The second view shared by neural network researchers is to base the learning of the synaptic matrix on a local Hebbian mechanism. The third assumption is the presence of chaos and the benefit gained by its presence. Chaos, although very simply produced, inherently possesses an infinite amount of cyclic regimes that can be exploited for coding information. Moreover, the network randomly wanders around these unstable regimes in a spontaneous way, thus rapidly proposing alternative responses to external stimuli, and being easily able to switch from one of these potential attractors to another in response to any incoming stimulus. Finally, since their introduction sixty years ago, cell assemblies have proved to be a powerful paradigm for brain information processing. After their introduction in artificial intelligence, cell assemblies became commonly used in computational neuroscience as a neural substrate for content addressable memories. <p> <p>Based on these assumptions, this thesis provides a computer model of neural network simulation of a brain-like memory. It first shows experimentally that the more information is to be stored in robust cyclic attractors, the more chaos appears as a regime in the background, erratically itinerating among brief appearances of these attractors. Chaos does not appear to be the cause, but the consequence of the learning. However, it appears as an helpful consequence that widens the network’s encoding capacity. To learn the information to be stored, two supervised iterative Hebbian learning algorithm are proposed. One leaves the semantics of the attractors to be associated with the feeding data unprescribed, while the other defines it a priori. Both algorithms show good results, even though the first one is more robust and has a greater storing capacity. Using these promising results, a biologically plausible alternative to these algorithms is proposed using cell assemblies as substrate for information. Even though this is not new, the mechanisms underlying their formation are poorly understood and, so far, there are no biologically plausible algorithms that can explain how external stimuli can be online stored in cell assemblies. This thesis provide such a solution combining a fast Hebbian/anti-Hebbian learning of the network's recurrent connections for the creation of new cell assemblies, and a slower feedback signal which stabilizes the cell assemblies by learning the feed forward input connections. This last mechanism is inspired by the retroaxonal hypothesis. <p> / Doctorat en Sciences / info:eu-repo/semantics/nonPublished

The diploma thesis is concerned with the analysis of recurrent word-combinations in the speech of advanced Czech speakers of English and native speakers of English. The data used for the analysis is extracted from two corpora, learner corpus LINDSEI and native speaker corpus LOCNEC. The aim of the thesis is to compare the two groups of speakers, determine differences in their use of recurrent word-combinations and compare the findings to previous studies involving speakers of different languages. The quantitative analysis is performed on a sample of 50 speakers from each corpus and the frequency data is used to compare the two groups as to the number of types of word-combinations they use and how frequently they do so. The qualitative analysis is performed on a sample of 15 speakers from each corpus to determine functional differences. Four categories of word-combinations are determined in the analysis. In the conclusion, the quantitative and qualitative findings are compared to previous research involving speakers of different languages. Keywords: spoken language, learner language, n-grams, n-gram analysis, recurrent word- combinations, lexical bundles, learner corpus

Le nombre croissant d’essais cliniques pour le traitement du cancer a conduit à la standardisation de l’évaluation de la réponse tumorale. Dans les essais cliniques de phase III des cancers avancés, la survie sans progression est souvent appliquée comme un critère de substitution pour la survie globale. Pour les tumeurs solides, la progression est généralement définie par les critères RECIST qui utilisent l’information sur le changement de taille des lésions cibles et les progressions de la maladie non-cible. Malgré leurs limites, les critères RECIST restent l’outil standard pour l’évaluation des traitements. En particulier, la taille tumorale mesurée au cours de temps est utilisée comme variable ponctuelle catégorisée pour identifier l’état d’un patient. L’approche statistique de la modélisation conjointe permet une analyse plus précise des marqueurs de réponse tumorale et de la survie. En outre, les modèles conjoints sont utiles pour les prédictions dynamiques individuelles. Dans ce travail, nous avons proposé d’appliquer un modèle conjoint trivarié pour des données longitudinales (taille tumorale), des évènements récurrents (les progressions de la maladie non-cible) et la survie. En utilisant des mesures de capacité prédictive, nous avons comparé le modèle proposé avec un modèle pour les progressions tumorales, définies selon les critères standards et la survie. Pour un essai clinique randomisé porté sur le cancer colorectal, nous avons trouvé une meilleure capacité prédictive du modèle proposé. Dans la deuxième partie, nous avons développé un logiciel en libre accès pour l’application de l’approche de modélisation conjointe proposée et les prédictions. Enfin, nous avons étendu le modèle à une analyse plus sophistiquée de l’évolution tumorale à l’aide d’un modèle mécaniste. Une équation différentielle ordinaire a été mise en œuvre pour décrire la trajectoire du marqueur biologique en tenant compte les caractéristiques biologiques de la croissance tumorale. Cette nouvelle approche contribue à la recherche clinique sur l’évaluation d’un traitement dans les essais cliniques grâce à une meilleure compréhension de la relation entre la réponse tumorale et la survie. / The increasing number of clinical trials for cancer treatments has led to standardization of guidelines for evaluation of tumor response. In phase III clinical trials of advanced cancer, progression-free survival is often applied as a surrogate endpoint for overall survival (OS). For solid tumors, progression is usually defined using the RECIST criteria that use information on the change of size of target lesions and progressions of non-target disease. The criteria remain the standard tool for treatment evaluation despite their limitations. In particular, repeatedly measured tumor size is used as a pointwise categorized variable to identify a patient’s status. Statistical approach of joint modeling allows for more accurate analysis of the tumor response markers and survival. Moreover, joint models are useful for individual dynamic predictions of death using patient’s history. In this work, we proposed to apply a trivariate joint model for a longitudinal outcome (tumor size), recurrent events (progressions of non-target disease) and survival. Using adapted measures of predictive accuracy we compared the proposed joint model with a model that considered tumor progressions defined within standard criteria and OS. For a randomized clinical trial for colorectal cancer patients, we found better predictive accuracy of the proposed joint model. In the second part, we developed freely available software for application of the proposed joint modeling and dynamic predictions approach. Finally, we extended the model to a more sophisticated analysis of tumor size evolution using a mechanistic model. An ordinary differential equation was implemented to describe the trajectory of the biomarker regarding the biological characteristics of tumor size under a treatment. This new approach contributes to clinical research on treatment evaluation in clinical trials by better understanding of the relationship between the markers of tumor response with OS.

The present work focuses on the two areas of investigation: system identification of helicopter and design of controller for the helicopter. Helicopter system identification, the first subject of investigation in this thesis, can be described as the extraction of system characteristics/dynamics from measured flight test data. Wind tunnel experimental data suffers from scale effects and model deficiencies. The increasing need for accurate models for the design of high bandwidth control system for helicopters has initiated a renewed interest in and a more active use of system identification. Besides, system identification is likely to become mandatory in the future for model validation of ground based helicopter simulators. Such simulators require accurate models in order to be accepted by pilots and regulatory authorities like Federal Aviation Regulation for realistic complementary helicopter mission training. Two approaches are widely used for system identification, namely, black box and gray box approach. In the black-box approach, the relationship between input-output data is approximated using nonparametric methods such as neural networks and in such a case, internal details of the system and model structure may not be known. In the gray box approach, parameters are estimated after defining the model structure. In this thesis, both black box and gray box approaches are investigated. In the black box approach, in this thesis, a comparative study and analysis of different Recurrent Neural Networks(RNN) for the identification of helicopter dynamics using flight data is investigated. Three different RNN architectures namely, Nonlinear Auto Regressive eXogenous input(NARX) model, neural network with internal memory known as Memory Neuron Networks(MNN)and Recurrent MultiLayer perceptron (RMLP) networks are used to identify dynamics of the helicopter at various flight conditions. Based on the results, the practical utility, advantages and limitations of the three models are critically appraised and it is found that the NARX model is most suitable for the identification of helicopter dynamics. In the gray box approach, helicopter model parameters are estimated after defining the model structure. The identification process becomes more difficult as the number of degrees-of-freedom and model parameters increase. To avoid the drawbacks of conventional methods, neural network based techniques, called the delta method is investigated in this thesis. This method does not require initial estimates of the parameters and the parameters can be directly extracted from the flight data. The Radial Basis Function Network(RBFN)is used for the purpose of estimation of parameters. It is shown that RBFN is able to satisfactorily estimate stability and control derivatives using the delta method. The second area of investigation addressed in this thesis is the control of helicopter in flight. Helicopter requires use of a control system to achieve satisfactory flight. Designing a classical controller involves developing a nonlinear model of the helicopter and extracting linearized state space matrices from the nonlinear model at various flight conditions. After examining the stability characteristics of the helicopter, the desired response is obtained using a feedback control system. The scheduling of controller gains over the entire envelope is used to obtain the desired response. In the present work, a helicopter having a soft inplane four bladed hingeless main rotor and a four-bladed tail rotor with conventional mechanical controls is considered. For this helicopter, a mathematical model and also a model based on neural network (using flight data) has been developed. As a precursor, a feed back controller, the Stability Augmentation System(SAS), is designed using linear quadratic regulator control with full state feedback and LQR with out put feedback approaches. SAS is designed to meet the handling qualities specification known as Aeronautical Design Standard ADS-33E-PRF. The control gains have been tuned with respect to forward speed and gain scheduling has been arrived at. The SAS in the longitudinal axis meets the requirement of the Level1 handling quality specifications in hover and low speed as well as for forward speed flight conditions. The SAS in the lateral axis meets the requirement of the Level2 handling quality specifications in both hover and low speed as well as for forward speed flight conditions. Such conventional design of control has served useful purposes, however, it requires considerable flight testing which is time consuming, to demonstrate and tune these control law gains. In modern helicopters, the stringent requirements and non-linear maneuvers make the controller design further complicated. Hence, new design tools have to be explored to control such helicopters. Among the many approaches in adaptive control, neural networks present a potential alternative for modeling and control of nonlinear dynamical systems due to their approximating capabilities and inherent adaptive features. Furthermore, from a practical perspective, the massive parallelism and fast adaptability of neural network implementations provide more incentive for further investigation in problems involving dynamical systems with unknown non-linearity. Therefore, adaptive control approach based on neural networks is proposed in this thesis. A neural network based Feedback Error Neural adaptive Controller(FENC) is designed for a helicopter. The proposed controller scheme is based on feedback error learning strategy in which the outer loop neural controller enhances the inner loop conventional controller by compensating for unknown non-linearity and parameter un-certainties. Nonlinear Auto Regressive eXogenous input(NARX)neural network architecture is used to approximate the control law and the controller network parameters are adapted using updated rules Lyapunov synthesis. An offline (finite time interval)and on-line adaptation strategy is used to approximate system uncertainties. The results are validated using simulation studies on helicopter undergoing an agile maneuver. The study shows that the neuro-controller meets the requirements of ADS-33 handling quality specifications. Even though the tracking error is less in FENC scheme, the control effort required to follow the command is very high. To overcome these problems, a Direct Adaptive Neural Control(DANC)scheme to track the rate command signal is presented. The neural controller is designed to track rate command signal generated using the reference model. For the simulation study, a linearized helicopter model at different straight and level flight conditions is considered. A neural network with a linear filter architecture trained using back propagation through time is used to approximate the control law. The controller network parameters are adapted using updated rules Lyapunov synthesis. The off-line trained (for finite time interval)network provides the necessary stability and tracking performance. The on-line learning is used to adapt the network under varying flight conditions. The on-line learning ability is demonstrated through parameter uncertainties. The performance of the proposed direct adaptive neural controller is compared with feedback error learning neural controller. The performance of the controller has been validated at various flight conditions. The theoretical results are validated using simulation studies based on a nonlinear six degree-of-freedom helicopter undergoing an agile maneuver. Realistic gust and sensor noise are added to the system to study the disturbance rejection properties of the neural controllers. To investigate the on-line learning ability of the proposed neural controller, different fault scenarios representing large model error and control surface loss are considered. The performances of the proposed DANC scheme is compared with the FENC scheme. The study shows that the neuro-controller meets the requirements of ADS-33 handling quality specifications.

For most of the naturally occurring signals, especially biomedical signals, the underlying physical process generating the signal is often not fully known, making it difficult to obtain a parametric model. Therefore, signal processing techniques are used to analyze the signal for non-parametrically characterizing the underlying system from which the signals are produced. Most of the real life systems are nonlinear and time varying, which poses a challenge while characterizing them. Additionally, multiple sensors are used to extract signals from such systems, resulting in multichannel signals which are inherently coupled. In this thesis, we counter this challenge by using Recurrence Plot based techniques for characterizing biomedical systems such as heart or brain, using signals such as heart rate variability (HRV), electroencephalogram(EEG) or functional magnetic resonance imaging (fMRI), respectively, extracted from them. In time series analysis, it is well known that a system can be represented by a trajectory in an N-dimensional state space, which completely represents an instance of the system behavior. Such a system characterization has been done using dynamical invariants such as correlation dimension, Lyapunov exponent etc. Takens has shown that when the state variables of the underlying system are not known, one can obtain a trajectory in ‘phase space’ using only the signals obtained from such a system. The phase space trajectory is topologically equivalent to the state space trajectory. This enables us to characterize the system behavior from only the signals sensed from them. However, estimation of correlation dimension, Lyapunov exponent, etc, are vulnerable to non-stationarities in the signal and require large number of sample points for accurate computation, both of which are important in the case of biomedical signals. Alternatively, a technique called Recurrence Plots (RP) has been proposed, which addresses these concerns, apart from providing additional insights. Measures to characterize RPs of single and two channel data are called Recurrence Quantification Analysis (RQA) and cross RQA (CRQA), respectively. These methods have been applied with a good measure of success in diverse areas. However, they have not been studied extensively in the context of experimental biomedical signals, especially multichannel data. In this thesis, the RP technique and its associated measures are briefly reviewed. Using the computational tools developed for this thesis, RP technique has been applied on select single channel, multichannel and multimodal (i.e. multiple channels derived from different modalities) biomedical signals. Connectivity analysis is demonstrated as post-processing of RP analysis on multichannel signals such as EEG and fMRI. Finally, a novel metric, based on the modification of a CRQA measure is proposed, which shows improved results. For the case of single channel signal, we have considered a large database of HRV signals of 112 subjects recorded for both normal and abnormal (anxiety disorder and depression disorder) subjects, in both supine and standing positions. Existing RQA measures, Recurrence Rate and Determinism, were used to distinguish between normal and abnormal subjects with an accuracy of 58.93%. A new measure, MLV has been introduced, using which a classification accuracy of 98.2% is obtained. Correlation between probabilities of recurrence (CPR) is a CRQA measure used to characterize phase synchronization between two signals. In this work, we demonstrate its utility with application to multimodal and multichannel biomedical signals. First, for the multimodal case, we have computed running CPR (rCPR), a modification proposed by us, which allows dynamic estimation of CPR as a function of time, on multimodal cardiac signals (electrocardiogram and arterial blood pressure) and demonstrated that the method can clearly detect abnormalities (premature ventricular contractions); this has potential applications in cardiac care such as assisted automated diagnosis. Second, for the multichannel case, we have used 16 channel EEG signals recorded under various physiological states such as (i) global epileptic seizure and pre-seizure and (ii) focal epilepsy. CPR was computed pair-wise between the channels and a CPR matrix of all pairs was formed. Contour plot of the CPR matrix was obtained to illustrate synchronization. Statistical analysis of CPR matrix for 16 subjects of global epilepsy showed clear differences between pre-seizure and seizure conditions, and a linear discriminant classifier was used in distinguishing between the two conditions with 100% accuracy. Connectivity analysis of multichannel EEG signals was performed by post-processing of the CPR matrix to understand global network-level characterization of the brain. Brain connectivity using thresholded CPR matrix of multichannel EEG signals showed clear differences in the number and pattern of connections in brain connectivity graph between epileptic seizure and pre-seizure. Corresponding brain headmaps provide meaningful insights about synchronization in the brain in those states. K-means clustering of connectivity parameters of CPR and linear correlation obtained from global epileptic seizure and pre-seizure showed significantly larger cluster centroid distances for CPR as opposed to linear correlation, thereby demonstrating the efficacy of CPR. The headmap in the case of focal epilepsy clearly enables us to identify the focus of the epilepsy which provides certain diagnostic value. Connectivity analysis on multichannel fMRI signals was performed using CPR matrix and graph theoretic analysis. Adjacency matrix was obtained from CPR matrices after thresholding it using statistical significance tests. Graph theoretic analysis based on communicability was performed to obtain community structures for awake resting and anesthetic sedation states. Concurrent behavioral data showed memory impairment due to anesthesia. Given the fact that previous studies have implicated the hippocampus in memory function, the CPR results showing the hippocampus within the community in awake state and out of it in anesthesia state, demonstrated the biological plausibility of the CPR results. On the other hand, results from linear correlation were less biologically plausible. In biological systems, highly synchronized and desynchronized systems are of interest rather than moderately synchronized ones. However, CPR is approximately a monotonic function of synchronization and hence can assume values which indicate moderate synchronization. In order to emphasize high synchronization/ desynchronization and de-emphasize moderate synchronization, a new method of Correlation Synchronization Convergence Time (CSCT) is proposed. It is obtained using an iterative procedure involving the evaluation of CPR for successive autocorrelations until CPR converges to a chosen threshold. CSCT was evaluated for 16 channel EEG data and corresponding contour plots and histograms were obtained, which shows better discrimination between synchronized and asynchronized states compared to the conventional CPR. This thesis has demonstrated the efficacy of RP technique and associated measures in characterizing various classes of biomedical signals. The results obtained are corroborated by well known physiological facts, and they provide physiologically meaningful insights into the functioning of the underlying biological systems, with potential diagnostic value in healthcare.