Extraction automatique et visualisation des thèmes abordés dans des résumés de mémoires et de thèses en anthropologie au Québec, de 1985 à 2009

S’insérant dans les domaines de la Lecture et de l’Analyse de Textes Assistées par Ordinateur (LATAO), de la Gestion Électronique des Documents (GÉD), de la visualisation de l’information et, en partie, de l’anthropologie, cette recherche exploratoire propose l’expérimentation d’une méthodologie descriptive en fouille de textes afin de cartographier thématiquement un corpus de textes anthropologiques. Plus précisément, nous souhaitons éprouver la méthode de classification hiérarchique ascendante (CHA) pour extraire et analyser les thèmes issus de résumés de mémoires et de thèses octroyés de 1985 à 2009 (1240 résumés), par les départements d’anthropologie de l’Université de Montréal et de l’Université Laval, ainsi que le département d’histoire de l’Université Laval (pour les résumés archéologiques et ethnologiques). En première partie de mémoire, nous présentons notre cadre théorique, c'est-à-dire que nous expliquons ce qu’est la fouille de textes, ses origines, ses applications, les étapes méthodologiques puis, nous complétons avec une revue des principales publications. La deuxième partie est consacrée au cadre méthodologique et ainsi, nous abordons les différentes étapes par lesquelles ce projet fut conduit; la collecte des données, le filtrage linguistique, la classification automatique, pour en nommer que quelques-unes. Finalement, en dernière partie, nous présentons les résultats de notre recherche, en nous attardant plus particulièrement sur deux expérimentations. Nous abordons également la navigation thématique et les approches conceptuelles en thématisation, par exemple, en anthropologie, la dichotomie culture ̸ biologie. Nous terminons avec les limites de ce projet et les pistes d’intérêts pour de futures recherches.

Improving sampling, optimization and feature extraction in Boltzmann machines

L’apprentissage supervisé de réseaux hiérarchiques à grande échelle connaît présentement un succès fulgurant. Malgré cette effervescence, l’apprentissage non-supervisé représente toujours, selon plusieurs chercheurs, un élément clé de l’Intelligence Artificielle, où les agents doivent apprendre à partir d’un nombre potentiellement limité de données. Cette thèse s’inscrit dans cette pensée et aborde divers sujets de recherche liés au problème d’estimation de densité par l’entremise des machines de Boltzmann (BM), modèles graphiques probabilistes au coeur de l’apprentissage profond. Nos contributions touchent les domaines de l’échantillonnage, l’estimation de fonctions de partition, l’optimisation ainsi que l’apprentissage de représentations invariantes. Cette thèse débute par l’exposition d’un nouvel algorithme d'échantillonnage adaptatif, qui ajuste (de fa ̧con automatique) la température des chaînes de Markov sous simulation, afin de maintenir une vitesse de convergence élevée tout au long de l’apprentissage. Lorsqu’utilisé dans le contexte de l’apprentissage par maximum de vraisemblance stochastique (SML), notre algorithme engendre une robustesse accrue face à la sélection du taux d’apprentissage, ainsi qu’une meilleure vitesse de convergence. Nos résultats sont présent ́es dans le domaine des BMs, mais la méthode est générale et applicable à l’apprentissage de tout modèle probabiliste exploitant l’échantillonnage par chaînes de Markov. Tandis que le gradient du maximum de vraisemblance peut-être approximé par échantillonnage, l’évaluation de la log-vraisemblance nécessite un estimé de la fonction de partition. Contrairement aux approches traditionnelles qui considèrent un modèle donné comme une boîte noire, nous proposons plutôt d’exploiter la dynamique de l’apprentissage en estimant les changements successifs de log-partition encourus à chaque mise à jour des paramètres. Le problème d’estimation est reformulé comme un problème d’inférence similaire au filtre de Kalman, mais sur un graphe bi-dimensionnel, où les dimensions correspondent aux axes du temps et au paramètre de température. Sur le thème de l’optimisation, nous présentons également un algorithme permettant d’appliquer, de manière efficace, le gradient naturel à des machines de Boltzmann comportant des milliers d’unités. Jusqu’à présent, son adoption était limitée par son haut coût computationel ainsi que sa demande en mémoire. Notre algorithme, Metric-Free Natural Gradient (MFNG), permet d’éviter le calcul explicite de la matrice d’information de Fisher (et son inverse) en exploitant un solveur linéaire combiné à un produit matrice-vecteur efficace. L’algorithme est prometteur: en terme du nombre d’évaluations de fonctions, MFNG converge plus rapidement que SML. Son implémentation demeure malheureusement inefficace en temps de calcul. Ces travaux explorent également les mécanismes sous-jacents à l’apprentissage de représentations invariantes. À cette fin, nous utilisons la famille de machines de Boltzmann restreintes “spike & slab” (ssRBM), que nous modifions afin de pouvoir modéliser des distributions binaires et parcimonieuses. Les variables latentes binaires de la ssRBM peuvent être rendues invariantes à un sous-espace vectoriel, en associant à chacune d’elles, un vecteur de variables latentes continues (dénommées “slabs”). Ceci se traduit par une invariance accrue au niveau de la représentation et un meilleur taux de classification lorsque peu de données étiquetées sont disponibles. Nous terminons cette thèse sur un sujet ambitieux: l’apprentissage de représentations pouvant séparer les facteurs de variations présents dans le signal d’entrée. Nous proposons une solution à base de ssRBM bilinéaire (avec deux groupes de facteurs latents) et formulons le problème comme l’un de “pooling” dans des sous-espaces vectoriels complémentaires.

Extraction of Glucan from Acremonium Diospyri and Its Application in Macrobrachium Rosenbergii Larval Rearing System Along With Bacterians as Microspheres

Present work is aimed at development of an appropriate microbial technology for protection of larvae of macrobrachium rosenbergii from disease and to increase survival rate in hatcheries. Application of immunostimulants to activate the immune system of cultured animals against pathogen is the widely accepted alternative to antibiotics in aquaculture. The most important immunostimulant is glucan. Therefore a research programme entitled as extraction of glucan from Acremonium diospyri and its application in macrobrachium rosenbergii larval rearing system along with bacterians as microspheres. The main objectives of the study are development of aquaculture grade glucan from acremonium diospyri, microencapsulated drug delivery system for the larvae of M. rosenbergii and microencapsulated glucan with bacterian preparation for the enhanced production of M. rosenbergii in larval rearing system. Based on the results of field trials microencapsulated glucan with bacterin preparation, it is concluded that the microencapsulated preparation at a concentration of 25g per million larvae once in seven days will enhance the production and quality seed of M. rosenbergii.

INVESTIGATIONS ON THE SOLVENT EXTRACTION AND LUMINESCENCE OF LANTHANOIDS WITH MIXTURES OF HETEROCYCLIC β- DIKETONE S AND VARIOUS NEUTRAL OXO-DONORS

The thesis entitled “ Investigations on the solvent extraction and luminescence of lanthanoids with mixtures of heterocyclic β-diketone S and various neutral oxo-donors” embodies the results of investigations carried out on the solvent extraction of trivalent lanthanoids with various heterocyclic β-diketones in the presence and absence of neutral oxo-donors and also on the luminescent studies of Eu3+-heterocyclic β-diketonate complexes with Lewis bases. The primary objective of the present work is to generate the knowledge base, especially to understand the interactions of lanthanoid-heterocyclic β-diketonates with various macrocyclic ligands such as crown ethers and neutral organophosphorus extractants , with a view to achieve better selectivity. The secondary objective of this thesis is to develop novel lanthanoid luminescent materials based on 3-phenyl-4-aroyl-5-isoxazolones and organophosphorus ligands, for use in electroluminescent devices. In the beginning it describes the need for the development of new mixed-ligand systems for the separation of lanthanoids and the development and importance of novel luminescent lanthanoid- β-diketonate complexes for display devices. The syntheses of various para substituted derivatives of 4-aroyl-5-isoxazolones and their characterization by various spectroscopic techniques are described. It also investigate the solvent extraction behaviour of trivalent lanthanoids with 4-aroyl-5-isoxazolones in the presence and absence of various crown ethers such as 18C6, DC18C6, DB18C6 and B18C6. Elemental analysis, IR and H NMR spectral studies are used to understand the interactions of crown ethers with 4-aroyl-5-isoxazolonate complexes of lanthanoids. The synergistic extraction of trivalent lanthanoids with sterically hindered 1-phenyl-3-methyl-4-pivaloyl-5-pyrazolone in the presence of various structurally related crown ethers are studied. The syntheses, characterization and photyphysical properties of Eu3+-4-aroyl-5-isoxazolonate complexes in the presence of Lewis bases like trictylphosphine oxide or triphenylphosphine oxide were studied.

Synergistic solvent extraction of Thorium(IV) and Uranium(VI) with β-diketones in presence of oxo-donors

The thesis entitled “Synergistic solvent extraction of Thorium(IV) and Uranium(VI) with β-diketones in presence of oxo-donors” embodies the results of the investigations carried out on the extraction of thorium(IV) an uranium(VI) with heterocyclic β-diketones in the presence and absence of various macrocyclic ligands and neutral organophosphorus extractants. The objective of this work is to generate the knowledge base to achieve better selectivity between thorium(IV) and uranium(VI) by understanding the interactions of crown ethers or neutral organophosphorus extractants with metal-heterocyclic β-diketonate complexes. Para-substituted 1-phenyl-3-methyl-4-aroyl-5-pyrazolones, namely,1-phenyl-3-methyl-4-(4-fluorobenzoyl)-5-pyrazolone (HPMFBP) and 1-phenyl-3-methyl-4-(4-toluoyl)-5-pyrazolone (HPMTP) were synthesized and characterized by elemental analysis, IR and H NMR spectral data. The synthesized ligands have been utilized for the extraction of thorium(IV) and uranium(VI) from nitric acid solutions in the presence and absence of various crown ethers. Thorium(IV) and uranium(VI) complexes with HPMPP(1-Phenyl-3-methyl-4-pivaloyl-5-pyrazolone) and neutral organophosphorus extractants were synthesized and characterized by IR and P NMR spectral data to further understand the interactions of neutral organophosphorus extractants with metal-chelates. Solid complexes of thorium(IV) and uranium(VI) with para-substituted 4-aroyl-5-isoxazolones and crown ethers were isolated and characterized by various spectroscopic techniques to further clarify the nature of the extracted complexes.

OFFLINE AND ONLINE SOLID PHASE EXTRACTION/PRECONCENTRATION OF INORGANICS

Solid phase extraction (SPE) is a powerful technique for preconcentration/removal or separation of trace and ultra trace amounts of toxic and nutrient elements. SPE effectively simplifies the labour intensive sample preparation, increase its reliability and eliminate the clean up step by using more selective extraction procedures. The synthesis of sorbents with a simplified procedure and diminution of the risks of errors shows the interest in the areas of environmental monitoring, geochemical exploration, food, agricultural, pharmaceutical, biochemical industry and high purity metal designing, etc. There is no universal SPE method because the sample pretreatment depends strongly on the analytical demand. But there is always an increasing demand for more sensitive, selective, rapid and reliable analytical procedures. Among the various materials, chelate modified naphthalene, activated carbon and chelate functionalized highly cross linked polymers are most important. In the biological and environmental field, large numbers of samples are to be analysed within a short span of time. Hence, online flow injection methods are preferred as they allow extraction, separation, identification and quantification of many numbers of analytes. The flow injection online preconcentration flame AAS procedure developed allows the determination of as low as 0.1 µg/l of nickel in soil and cobalt in human hair samples. The developed procedure is precise and rapid and allows the analysis of 30 samples per hour with a loading time of 60 s. The online FI manifold used in the present study permits high sampling, loading rates and thus resulting in higher preconcentration/enrichment factors of -725 and 600 for cobalt and nickel respectively with a 1 min preconcentration time compared to conventional FAAS signal. These enrichment factors are far superior to hitherto developed on line preconcentration procedures for inorganics. The instrumentation adopted in the present study allows much simpler equipment and low maintenance costs compared to costlier ICP-AES or ICP-MS instruments.

Statistical signal extraction using stable processes

The standard models for statistical signal extraction assume that the signal and noise are generated by linear Gaussian processes. The optimum filter weights for those models are derived using the method of minimum mean square error. In the present work we study the properties of signal extraction models under the assumption that signal/noise are generated by symmetric stable processes. The optimum filter is obtained by the method of minimum dispersion. The performance of the new filter is compared with their Gaussian counterparts by simulation.

Optimization of Process Parameters for the Extraction of Gelatin from the Skin of Freshwater Fish and the Evaluation of Physical anal Chemical Characteristics

Faculty of Marine Sciences,Cochin University of Science and Technology

“Improved Feature Extraction and Classification Techniques for Multispectral Brain Magnetic Resonance Images”

Magnetic Resonance Imaging (MRI) is a multi sequence medical imaging technique in which stacks of images are acquired with different tissue contrasts. Simultaneous observation and quantitative analysis of normal brain tissues and small abnormalities from these large numbers of different sequences is a great challenge in clinical applications. Multispectral MRI analysis can simplify the job considerably by combining unlimited number of available co-registered sequences in a single suite. However, poor performance of the multispectral system with conventional image classification and segmentation methods makes it inappropriate for clinical analysis. Recent works in multispectral brain MRI analysis attempted to resolve this issue by improved feature extraction approaches, such as transform based methods, fuzzy approaches, algebraic techniques and so forth. Transform based feature extraction methods like Independent Component Analysis (ICA) and its extensions have been effectively used in recent studies to improve the performance of multispectral brain MRI analysis. However, these global transforms were found to be inefficient and inconsistent in identifying less frequently occurred features like small lesions, from large amount of MR data. The present thesis focuses on the improvement in ICA based feature extraction techniques to enhance the performance of multispectral brain MRI analysis. Methods using spectral clustering and wavelet transforms are proposed to resolve the inefficiency of ICA in identifying small abnormalities, and problems due to ICA over-completeness. Effectiveness of the new methods in brain tissue classification and segmentation is confirmed by a detailed quantitative and qualitative analysis with synthetic and clinical, normal and abnormal, data. In comparison to conventional classification techniques, proposed algorithms provide better performance in classification of normal brain tissues and significant small abnormalities.

Feature Extraction Methods Based on Linear Predictive Coding and Wavelet Packet Decomposition for Recognizing Spoken Words in Malayalam

Speech signals are one of the most important means of communication among the human beings. In this paper, a comparative study of two feature extraction techniques are carried out for recognizing speaker independent spoken isolated words. First one is a hybrid approach with Linear Predictive Coding (LPC) and Artificial Neural Networks (ANN) and the second method uses a combination of Wavelet Packet Decomposition (WPD) and Artificial Neural Networks. Voice signals are sampled directly from the microphone and then they are processed using these two techniques for extracting the features. Words from Malayalam, one of the four major Dravidian languages of southern India are chosen for recognition. Training, testing and pattern recognition are performed using Artificial Neural Networks. Back propagation method is used to train the ANN. The proposed method is implemented for 50 speakers uttering 20 isolated words each. Both the methods produce good recognition accuracy. But Wavelet Packet Decomposition is found to be more suitable for recognizing speech because of its multi-resolution characteristics and efficient time frequency localizations

COMBINED FEATURE EXTRACTION TECHNIQUES AND NAIVE BAYES CLASSIFIER FOR SPEECH RECOGNITION

Speech processing and consequent recognition are important areas of Digital Signal Processing since speech allows people to communicate more natu-rally and efficiently. In this work, a speech recognition system is developed for re-cognizing digits in Malayalam. For recognizing speech, features are to be ex-tracted from speech and hence feature extraction method plays an important role in speech recognition. Here, front end processing for extracting the features is per-formed using two wavelet based methods namely Discrete Wavelet Transforms (DWT) and Wavelet Packet Decomposition (WPD). Naive Bayes classifier is used for classification purpose. After classification using Naive Bayes classifier, DWT produced a recognition accuracy of 83.5% and WPD produced an accuracy of 80.7%. This paper is intended to devise a new feature extraction method which produces improvements in the recognition accuracy. So, a new method called Dis-crete Wavelet Packet Decomposition (DWPD) is introduced which utilizes the hy-brid features of both DWT and WPD. The performance of this new approach is evaluated and it produced an improved recognition accuracy of 86.2% along with Naive Bayes classifier.

A Comparative Study of Wavelet Based Feature Extraction Techniques in Recognizing Isolated Spoken Words

Speech is a natural mode of communication for people and speech recognition is an intensive area of research due to its versatile applications. This paper presents a comparative study of various feature extraction methods based on wavelets for recognizing isolated spoken words. Isolated words from Malayalam, one of the four major Dravidian languages of southern India are chosen for recognition. This work includes two speech recognition methods. First one is a hybrid approach with Discrete Wavelet Transforms and Artificial Neural Networks and the second method uses a combination of Wavelet Packet Decomposition and Artificial Neural Networks. Features are extracted by using Discrete Wavelet Transforms (DWT) and Wavelet Packet Decomposition (WPD). Training, testing and pattern recognition are performed using Artificial Neural Networks (ANN). The proposed method is implemented for 50 speakers uttering 20 isolated words each. The experimental results obtained show the efficiency of these techniques in recognizing speech