Дисертації: "Discriminative classifier"

1

Masip, Rodó David. "Face Classification Using Discriminative Features and Classifier Combination." Doctoral thesis, Universitat Autònoma de Barcelona, 2005. http://hdl.handle.net/10803/3051.

Повний текст джерела

Анотація:

A mesura que la tecnologia evoluciona, apareixen noves aplicacions en el mon de la classificació facial. En el reconeixement de patrons, normalment veiem les cares com a punts en un espai de alta dimensionalitat definit pels valors dels seus pixels. Aquesta aproximació pateix diversos problemes: el fenomen de la "la maledicció de la dimensionalitat", la presència d'oclusions parcials o canvis locals en la il·luminació. Tradicionalment, només les característiques internes de les imatges facials s'han utilitzat per a classificar, on normalment es fa una extracció de característiques. Les tècniques d'extracció de característiques permeten reduir la influencia dels problemes mencionats, reduint també el soroll inherent de les imatges naturals alhora que es poden aprendre característiques invariants de les imatges facials. En la primera part d'aquesta tesi presentem alguns mètodes d'extracció de característiques clàssics: Anàlisi de Components Principals (PCA), Anàlisi de Components Independents (ICA), Factorització No Negativa de Matrius (NMF), i l'Anàlisi Discriminant de Fisher (FLD), totes elles fent alguna mena d'assumpció en les dades a classificar. La principal contribució d'aquest treball es una nova família de tècniques d'extracció de característiques usant el algorisme del Adaboost. El nostre mètode no fa cap assumpció en les dades a classificar, i construeix de forma incremental la matriu de projecció tenint en compte els exemples mes difícils Per altra banda, en la segon apart de la tesi explorem el rol de les característiques externes en el procés de classificació facial, i presentem un nou mètode per extreure un conjunt alineat de característiques a partir de la informació externa que poden ser combinades amb les tècniques clàssiques millorant els resultats globals de classificació. As technology evolves, new applications dealing with face classification appear. In pattern recognition, faces are usually seen as points in a high dimensional spaces defined by their pixel values. This approach must deal with several problems such as: the curse of dimensionality, the presence of partial occlusions or local changes in the illumination. Traditionally, only the internal features of face images have been used for classification purposes, where usually a feature extraction step is performed. Feature extraction techniques allow to reduce the influence of the problems mentioned, reducing also the noise inherent from natural images and learning invariant characteristics from face images. In the first part of this thesis some internal feature extraction methods are presented: Principal Component Analysis (PCA), Independent Component Analysis (ICA), Non Negative Matrix Factorization (NMF), and Fisher Linear Discriminant Analysis (FLD), all of them making some kind of the assumption on the data to classify. The main contribution of our work is a non parametric feature extraction family of techniques using the Adaboost algorithm. Our method makes no assumptions on the data to classify, and incrementally builds the projection matrix taking into account the most difficult samples. On the other hand, in the second part of this thesis we also explore the role of external features in face classification purposes, and present a method for extracting an aligned feature set from external face information that can be combined with the classic internal features improving the global performance of the face classification task.

Стилі APA, Harvard, Vancouver, ISO та ін.

2

Georgatzis, Konstantinos. "Dynamical probabilistic graphical models applied to physiological condition monitoring." Thesis, University of Edinburgh, 2017. http://hdl.handle.net/1842/28838.

Повний текст джерела

Анотація:

Intensive Care Units (ICUs) host patients in critical condition who are being monitored by sensors which measure their vital signs. These vital signs carry information about a patient’s physiology and can have a very rich structure at fine resolution levels. The task of analysing these biosignals for the purposes of monitoring a patient’s physiology is referred to as physiological condition monitoring. Physiological condition monitoring of patients in ICUs is of critical importance as their health is subject to a number of events of interest. For the purposes of this thesis, the overall task of physiological condition monitoring is decomposed into the sub-tasks of modelling a patient’s physiology a) under the effect of physiological or artifactual events and b) under the effect of drug administration. The first sub-task is concerned with modelling artifact (such as the taking of blood samples, suction events etc.), and physiological episodes (such as bradycardia), while the second sub-task is focussed on modelling the effect of drug administration on a patient’s physiology. The first contribution of this thesis is the formulation, development and validation of the Discriminative Switching Linear Dynamical System (DSLDS) for the first sub-task. The DSLDS is a discriminative model which identifies the state-of-health of a patient given their observed vital signs using a discriminative probabilistic classifier, and then infers their underlying physiological values conditioned on this status. It is demonstrated on two real-world datasets that the DSLDS is able to outperform an alternative, generative approach in most cases of interest, and that an a-mixture of the two models achieves higher performance than either of the two models separately. The second contribution of this thesis is the formulation, development and validation of the Input-Output Non-Linear Dynamical System (IO-NLDS) for the second sub-task. The IO-NLDS is a non-linear dynamical system for modelling the effect of drug infusions on the vital signs of patients. More specifically, in this thesis the focus is on modelling the effect of the widely used anaesthetic drug Propofol on a patient’s monitored depth of anaesthesia and haemodynamics. A comparison of the IO-NLDS with a model derived from the Pharmacokinetics/Pharmacodynamics (PK/PD) literature on a real-world dataset shows that significant improvements in predictive performance can be provided without requiring the incorporation of expert physiological knowledge.

Стилі APA, Harvard, Vancouver, ISO та ін.

3

Klautau, Aldebaro. "Speech recognition using discriminative classifiers /." Diss., Connect to a 24 p. preview or request complete full text in PDF format. Access restricted to UC campuses, 2003. http://wwwlib.umi.com/cr/ucsd/fullcit?p3091208.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

4

Xue, Jinghao. "Aspects of generative and discriminative classifiers." Thesis, Connect to e-thesis, 2008. http://theses.gla.ac.uk/272/.

Повний текст джерела

Анотація:

Thesis (Ph.D.) - University of Glasgow, 2008. Ph.D. thesis submitted to the Department of Statistics, Faculty of Information and Mathematical Sciences, University of Glasgow, 2008. Includes bibliographical references. Print version also available.

Стилі APA, Harvard, Vancouver, ISO та ін.

5

Pernot, Etienne. "Choix d'un classifieur en discrimination." Paris 9, 1994. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1994PA090014.

Повний текст джерела

Анотація:

Dans cette thèse, nous nous posons le problème de la détermination du classifieur le plus adapté à résoudre un problème donné de discrimination. Le choix du classifieur est déjà guidé par des contraintes opérationnelles, mais au-delà de ces contraintes, et après que le classifieur a été configuré grâce à une base d'apprentissage, c'est le taux de généralisation du classifieur (ou taux de réussite) qui est le critère caractérisant sa performance. Ce taux, généralement inconnu, est estimé à l'aide d'une base de généralisation. Cette estimation dépend donc du problème de discrimination étudié, du classifieur utilisé, de la base d'apprentissage et de la base de généralisation. Ces différentes dépendances sont étudiées soit théoriquement, soit de manière expérimentale, sur une douzaine de classifieurs différents, neuronaux et classiques. Le problème de la validité de la comparaison de deux classifieurs par les estimations de leur taux de généralisation est aussi étudié, et nous obtenons des informations sur les tailles relatives à donner aux bases d'apprentissage et de généralisation. Dans un objectif de comparaison de classifieurs, Neuroclasse, un outil logiciel donnant la possibilité de tester un grand nombre de classifieurs différents, a été développé, et est précisément décrit. Dans Neuroclasse est aussi intégré un système pour la détermination automatique du classifieur fournissant le meilleur taux de généralisation estimé sur une base de généralisation fixée. Ce système est implanté sous forme d'un système expert. Ce système, testé sur différentes bases de données, donne de bons résultats, mais met en évidence un phénomène d'apprentissage de la base de généralisation, dû aux tests successifs de nombreux classifieurs sur une même base de généralisation. Nous étudions ce phénomène expérimentalement, et nous donnons un ordre de grandeur du nombre de classifieurs qu'il est possible de tester en limitant cet effet

Стилі APA, Harvard, Vancouver, ISO та ін.

6

Katz, Marcel [Verfasser]. "Discriminative classifiers for speaker Recognition / Marcel Katz." Saarbrücken : Südwestdeutscher Verlag für Hochschulschriften, 2009. http://www.vdm-verlag.de.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

7

ABDALLAH, HICHAM. "Application de l'analyse relationnelle pour classifier descripteurs et modalites en mode discrimination." Paris 6, 1996. http://www.theses.fr/1996PA066001.

Повний текст джерела

Анотація:

Dans le processus presente dans la these, quatre niveaux fondamentaux ont ete a l'origine de notre travail: 1 la classification de l'espace des descripteurs ou des variables pour reduire et resserrer cet espace: en introduisant de nouvelles structures (par des criteres d'associations connus) permettant de donner des bornes ou seuils au dela desquels on considere les deux descripteurs ou variables sont ressemblants. Un point important ici est la possibilite de tout expliciter en fonction du rand ou du chi-deux. Cette etude permet de voir apparaitre la notion de severite pour chaque critere, qui nous permet de choisir suivant le contexte ceux qui sont les plus adaptes. 2 une fois trouvee la partition des descripteurs. On agrege les descripteurs de chaque classe par le descripteur consensus. 3 la classification des modalites pour la structuration et pour la phase de discrimination proprement dite ou le regroupement modalitaire permet une interpretation plus grande du phenomene de caracterisation des concepts. Ceci entre autre a ete obtenu apres une interrogation sur le bien fonde de la notion de referant pour les indices de similarite avec une caracterisation de leurs proprietes metriques. De meme, cette etude nous permet de choisir suivant le contexte les indices qui sont les plus adaptes. 4 le raffinement des concepts

Стилі APA, Harvard, Vancouver, ISO та ін.

8

Dastile, Xolani Collen. "Improved tree species discrimination at leaf level with hyperspectral data combining binary classifiers." Thesis, Rhodes University, 2011. http://hdl.handle.net/10962/d1002807.

Повний текст джерела

Анотація:

The purpose of the present thesis is to show that hyperspectral data can be used for discrimination between different tree species. The data set used in this study contains the hyperspectral measurements of leaves of seven savannah tree species. The data is high-dimensional and shows large within-class variability combined with small between-class variability which makes discrimination between the classes challenging. We employ two classification methods: G-nearest neighbour and feed-forward neural networks. For both methods, direct 7-class prediction results in high misclassification rates. However, binary classification works better. We constructed binary classifiers for all possible binary classification problems and combine them with Error Correcting Output Codes. We show especially that the use of 1-nearest neighbour binary classifiers results in no improvement compared to a direct 1-nearest neighbour 7-class predictor. In contrast to this negative result, the use of neural networks binary classifiers improves accuracy by 10% compared to a direct neural networks 7-class predictor, and error rates become acceptable. This can be further improved by choosing only suitable binary classifiers for combination.

Стилі APA, Harvard, Vancouver, ISO та ін.

9

Rüther, Johannes. "Navigating Deep Classifiers : A Geometric Study Of Connections Between Adversarial Examples And Discriminative Features In Deep Neural Networks." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-291775.

Повний текст джерела

Анотація:

Although deep networks are powerful and effective in numerous applications, their high vulnerability to adversarial perturbations remains a critical limitation in domains such as security, personalized medicine or autonomous systems. While the sensitivity to adversarial perturbations is generally viewed as a bug of deep classifiers, recent research suggests that they are actually a manifestation of non-robust features that deep classifiers exploit for predictive accuracy. In this work, we therefore systematically compute and analyze these perturbations to understand how they relate to discriminative features that models use. Most of the insights obtained in this work take a geometrical perspective on classifiers, specifically the location of decision boundaries in the vicinity of samples. Perturbations that successfully flip classification decisions are conceived as directions in which samples can be moved to transition into other classification regions. Thereby we reveal that navigating classification spaces is surprisingly simple: Any sample can be moved into a target region within a small distance by following a single direction extracted from adversarial perturbations. Moreover, we reveal that for simple data sets such as MNIST, discriminative features used by deep classifiers with standard training are indeed composed of elements found in adversarial examples. Finally, our results also demonstrate that adversarial training fundamentally changes classifier geometry in the vicinity of samples, yielding more diverse and complex decision boundaries. Även om djupa neurala nät är kraftfulla och effektiva i många användningar, är deras stora sårbarhet för medvetna störningar (adversarial perturbations) fortfarande en kritisk begränsning inom områden som säkerhet, individanpassad medicin eller autonoma system. Även om känsligheten för medvetna störningar i allmänhet betraktas som en brist hos klassifierare baserade på djupa nät, tyder färsk forskning på att de i själva verket är ett uttryck för orobusta features som klassifierarna utnyttjar för att göra exakta prediktioner. I detta arbete beräknar och analyserar vi därför systematiskt dessa störningar för att förstå hur de förhåller sig till diskriminativa features som modellerna använder. De flesta insikter som erhålls i detta arbete har ett geometriskt perspektiv på klassificerare, särskilt placeringen av beslutsgränserna i närheten av datasamplen. Störningar som framgångsrikt kan ändra på klassificeringsbeslut utformas som riktning där datasamplen kan flyttas in till andra klassificeringsregioner. På så sätt avslöjar vi att det är förvånansvärt enkelt att navigera i klassificeringsrymden: Ett godtyckligt sampel kan flyttas till en annan närliggande klassificeringsregion genom att man följer riktningen som extraherats från medvetna störningar. Dessutom avslöjar vi att när det gäller enkla datauppsättningar som MNIST, består de diskriminerande features som används av djupa klassifierare, tränade med standardmetoder, faktiskt av element som återfinns bland de medvetna störningsexemplen. Slutligen visar våra resultat också att medvetna störningar i grunden förändrar klassificerargeometrin i närheten av datasampel, vilket ger mer varierande och komplexa beslutsgränser.

Стилі APA, Harvard, Vancouver, ISO та ін.

10

Musayeva, Khadija. "Generalization Performance of Margin Multi-category Classifiers." Thesis, Université de Lorraine, 2019. http://www.theses.fr/2019LORR0096/document.

Повний текст джерела

Анотація:

Cette thèse porte sur la théorie de la discrimination multi-classe à marge. Elle a pour cadre la théorie statistique de l’apprentissage de Vapnik et Chervonenkis. L’objectif est d’établir des bornes de généralisation possédant une dépendances explicite au nombre C de catégories, à la taille m de l’échantillon et au paramètre de marge gamma, lorsque la fonction de perte considérée est une fonction de perte à marge possédant la propriété d’être lipschitzienne. La borne de généralisation repose sur la performance empirique du classifieur ainsi que sur sa "capacité". Dans cette thèse, les mesures de capacité considérées sont les suivantes : la complexité de Rademacher, les nombres de recouvrement et la dimension fat-shattering. Nos principales contributions sont obtenues sous l’hypothèse que les classes de fonctions composantes calculées par le classifieur ont des dimensions fat-shattering polynomiales et que les fonctions composantes sont indépendantes. Dans le contexte du schéma de calcul introduit par Mendelson, qui repose sur les relations entre les mesures de capacité évoquées plus haut, nous étudions l’impact que la décomposition au niveau de l’une de ces mesures de capacité a sur les dépendances (de la borne de généralisation) à C, m et gamma. En particulier, nous démontrons que la dépendance à C peut être considérablement améliorée par rapport à l’état de l’art si la décomposition est reportée au niveau du nombre de recouvrement ou de la dimension fat-shattering. Ce changement peut affecter négativement le taux de convergence (dépendance à m), ce qui souligne le fait que l’optimisation par rapport aux trois paramètres fondamentaux se traduit par la recherche d’un compromis This thesis deals with the theory of margin multi-category classification, and is based on the statistical learning theory founded by Vapnik and Chervonenkis. We are interested in deriving generalization bounds with explicit dependencies on the number C of categories, the sample size m and the margin parameter gamma, when the loss function considered is a Lipschitz continuous margin loss function. Generalization bounds rely on the empirical performance of the classifier as well as its "capacity". In this work, the following scale-sensitive capacity measures are considered: the Rademacher complexity, the covering numbers and the fat-shattering dimension. Our main contributions are obtained under the assumption that the classes of component functions implemented by a classifier have polynomially growing fat-shattering dimensions and that the component functions are independent. In the context of the pathway of Mendelson, which relates the Rademacher complexity to the covering numbers and the latter to the fat-shattering dimension, we study the impact that decomposing at the level of one of these capacity measures has on the dependencies on C, m and gamma. In particular, we demonstrate that the dependency on C can be substantially improved over the state of the art if the decomposition is postponed to the level of the metric entropy or the fat-shattering dimension. On the other hand, this impacts negatively the rate of convergence (dependency on m), an indication of the fact that optimizing the dependencies on the three basic parameters amounts to looking for a trade-off

Стилі APA, Harvard, Vancouver, ISO та ін.

11

Dang, Thanh Ha. "Mesures de discrimination et leurs applications en apprentissage inductif." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2007. http://tel.archives-ouvertes.fr/tel-00184691.

Повний текст джерела

Анотація:

De nos jours, les données disponibles deviennent de plus en plus volumineuses et elles peuvent être de nature très diverse : vagues, manquantes, numériques, symboliques par exemple. Or ce qui importe à l'utilisateur, ce ne sont pas les données elles-mêmes, mais les connaissances qu'on peut en extraire. Face à la quantité de données disponibles, le traitement efficace de données est problématique. Dans cette thèse, nous adoptons une approche d'extraction de connaissances à partir de données basée sur l'apprentissage inductif, plus précisément, par arbres de décision. De façon générale, un système construit par apprentissage inductif a pour but de discriminer les individus de différentes classes. Sa qualité dépend de la capacité de discrimination qu'il acquiert au cours de l'apprentissage au travers des données. En particulier, un algorithme de construction d'arbre de décision procède par évaluation successive de la capacité de discrimination des attributs pour construire l'arbre de décision. Nos travaux concernent l'étude des mesures de discrimination tant classiques que floues, et leurs applications en apprentissage inductif. D'une part, nous nous intéressons aux mesures de discrimination dans la construction des arbres de décision. Dans un premier temps, ces mesures font l'objet d'une étude selon une approche axiomatique. Nous développons un nouveau modèle pour caractériser les mesures de discriminations floues. Dans un deuxième temps, nous proposons d'utiliser ces mesures dans les différentes étapes de la construction des arbres de décision flous. D'autre part, nous étudions l'utilisation de ces mesures de discrimination pour d'autres aspects de l'apprentissage. Nous examinons tout d'abord le problème de l'évaluation des classifieurs et proposons une méthode basée sur l'utilisation de la notion de capacité de discrimination. Enfin, nous considérons le problème du traitement des données manquantes et proposons une technique de substitution des valeurs manquantes, qui restitue la capacité de discrimination des attributs. Ces travaux sont validés sur des données conventionnelles et appliqués à des données réelles dans le cadre de deux applications qui concernent la classification de courriers électroniques et la classification de traces d'interactions homme-machine.

Стилі APA, Harvard, Vancouver, ISO та ін.

12

Asper, Roman Yorick [Verfasser], Stephan [Akademischer Betreuer] Waack, and Carsten [Akademischer Betreuer] Damm. "Classifiers for Discrimination of Significant Protein Residues and Protein-Protein Interaction Using Concepts of Information Theory and Machine Learning / Roman Yorick Asper. Gutachter: Stephan Waack ; Carsten Damm. Betreuer: Stephan Waack." Göttingen : Niedersächsische Staats- und Universitätsbibliothek Göttingen, 2012. http://d-nb.info/1042969108/34.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

13

Gul, Ahmet Bahtiyar. "Holistic Face Recognition By Dimension Reduction." Master's thesis, METU, 2003. http://etd.lib.metu.edu.tr/upload/1056738/index.pdf.

Повний текст джерела

Анотація:

Face recognition is a popular research area where there are different approaches studied in the literature. In this thesis, a holistic Principal Component Analysis (PCA) based method, namely Eigenface method is studied in detail and three of the methods based on the Eigenface method are compared. These are the Bayesian PCA where Bayesian classifier is applied after dimension reduction with PCA, the Subspace Linear Discriminant Analysis (LDA) where LDA is applied after PCA and Eigenface where Nearest Mean Classifier applied after PCA. All the three methods are implemented on the Olivetti Research Laboratory (ORL) face database, the Face Recognition Technology (FERET) database and the CNN-TURK Speakers face database. The results are compared with respect to the effects of changes in illumination, pose and aging. Simulation results show that Subspace LDA and Bayesian PCA perform slightly well with respect to PCA under changes in pose however, even Subspace LDA and Bayesian PCA do not perform well under changes in illumination and aging although they perform better than PCA.

Стилі APA, Harvard, Vancouver, ISO та ін.

14

Didiot, Emmanuel. "Segmentation parole/musique pour la transcription automatique de parole continue." Phd thesis, Université Henri Poincaré - Nancy I, 2007. http://tel.archives-ouvertes.fr/tel-00187941.

Повний текст джерела

Анотація:

Dans cette thèse, nous étudions la segmentation d'un flux audio en parole, musique et parole sur musique (P/M). Cette étape est fondamentale pour toute application basée sur la transcription automatique de flux radiophoniques et plus généralement multimédias. L'application visée ici est un système de détection de mots clés dans les émissions radiophoniques. Les performances de ce système dépendront de la bonne segmentation du signal fournie par le système de discrimination parole/musique. En effet, une mauvaise classification du signal peut provoquer des omissions de mots clés ou des fausses alarmes. Afin d'améliorer la discrimination parole/musique, nous proposons une nouvelle méthode de paramétrisation du signal. Nous utilisons la décomposition en ondelettes qui permet une analyse des signaux non stationnaires dont la musique est un exemple. Nous calculons différentes énergies sur les coefficients d'ondelettes pour construire nos vecteurs de paramètres. Le signal est alors segmenté en quatre classes : parole (P), non-parole (NP), musique (M) et non-musique (NM) grâce à deux systèmes disjoints de classification HMM classe/non-classe. Cette architecture a été choisie car elle permet de trouver les meilleurs paramètres indépendamment pour chaque tâche P/NP et M/NM. Une fusion des sorties des classifieurs est alors effectuée pour obtenir la décision finale : parole, musique ou parole sur musique. Les résultats obtenus sur un corpus réel d'émissions de radio montrent que notre paramétrisation en ondelettes apporte une nette amélioration des performances en discrimination M/NM et P/M par rapport à la paramétrisation de référence fondée sur les coefficients cepstraux.

Стилі APA, Harvard, Vancouver, ISO та ін.

15

Sun, Yong-Peng. "A Discriminative Locally-Adaptive Nearest Centroid Classifier for Phoneme Classification." Thesis, 2012. http://hdl.handle.net/10012/6968.

Повний текст джерела

Анотація:

Phoneme classification is a key area of speech recognition. Phonemes are the basic modeling units in modern speech recognition and they are the constructive units of words. Thus, being able to quickly and accurately classify phonemes that are input to a speech-recognition system is a basic and important step towards improving and eventually perfecting speech recognition as a whole. Many classification approaches currently exist that can be applied to the task of classifying phonemes. These techniques range from simple ones such as the nearest centroid classifier to complex ones such as support vector machine. Amongst the existing classifiers, the simpler ones tend to be quicker to train but have lower accuracy, whereas the more complex ones tend to be higher in accuracy but are slower to train. Because phoneme classification involves very large datasets, it is desirable to have classifiers that are both quick to train and are high in accuracy. The formulation of such classifiers is still an active ongoing research topic in phoneme classification. One paradigm in formulating such classifiers attempts to increase the accuracies of the simpler classifiers with minimal sacrifice to their running times. The opposite paradigm attempts to increase the training speeds of the more complex classifiers with minimal sacrifice to their accuracies. The objective of this research is to develop a new centroid-based classifier that builds upon the simpler nearest centroid classifier by incorporating a new discriminative locally-adaptive training procedure developed from recent advances in machine learning. This new classifier, which is referred to as the discriminative locally-adaptive nearest centroid (DLANC) classifier, achieves much higher accuracies as compared to the nearest centroid classifier whilst having a relatively low computational complexity and being able to scale up to very large datasets.

Стилі APA, Harvard, Vancouver, ISO та ін.

16

Hoang, Tran Vu, and 陳武黃. "A Study of Parking Space Detection Based On a Multi-Layer Discriminative Classifier." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/y75ky9.

Повний текст джерела

Анотація:

碩士 國立高雄應用科技大學 製造與管理外國學生碩士專班 103 In this research, we proposed a novel semantic inference framework with multiple layers for vacant parking space detection. The framework consists of an image layer, a patch layer, a space layer, and a lot layer. In the image layer, image patches are selected based on the 3-D parking lot structure. We found the occlusion pattern within each patch reveals partial cues of parking status. Thus, our system extracted lighting-invariant features of patches and trained weak classifiers to recognize the occlusion pattern in the patch layer. The outputs of the classifiers, presenting the types of inter-object occlusion, were treated as the mid-level features and inputted to the space layer. Next, a boosted space classifier was trained to recognize the mid-level features and output the status of a 3-space unit in a probability fashion. In the lot layer, we regarded these local status decisions as high-level evidences and proposed a Markov Random Field to infer the final status of the parking lot. Our results show that the proposed framework can overcome the inter-object occlusion and achieve better space detection in different weather conditions.

Стилі APA, Harvard, Vancouver, ISO та ін.

17

Katz, Marcel [Verfasser]. "Discriminative classifiers for speaker recognition / von Marcel Katz." 2008. http://d-nb.info/989980413/34.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

18

Binder, Carolyn. "Using an Aural Classifier to Discriminate Cetacean Vocalizations." 2012. http://hdl.handle.net/10222/14607.

Повний текст джерела

Анотація:

To positively identify marine mammals using passive acoustics, large volumes of data are often collected that need to be processed by a trained analyst. To reduce acoustic analyst workload, an automatic detector can be implemented that produces many detections, which feed into an automatic classifier to significantly reduce the number of false detections. This requires the development of a robust classifier capable of performing inter-species classification as well as discriminating cetacean vocalizations from anthropogenic noise sources. A prototype aural classifier was developed at Defence Research and Development Canada that uses perceptual signal features which model the features employed by the human auditory system. The dataset included anthropogenic passive transients and vocalizations from five cetacean species: bowhead, humpback, North Atlantic right, minke and sperm whales. Discriminant analysis was implemented to replace principal component analysis; the projection obtained using discriminant analysis improved between-species discrimination during multiclass cetacean classification, compared to principal component analysis. The aural classifier was able to successfully identify the vocalizing cetacean species. The area under the receiver operating characteristic curve (AUC) is used to quantify the two-class classifier performance and the M-measure is used when there are three or more classes; the maximum possible value of both AUC and M is 1.00 – which is indicative of an ideal classifier model. Accurate classification results were obtained for multiclass classification of all species in the dataset (M = 0.99), and the challenging bowhead/ humpback (AUC = 0.97) and sperm whale click/anthropogenic transient (AUC = 1.00) two-class classifications.

Стилі APA, Harvard, Vancouver, ISO та ін.

19

Liu, Pengfei. "Discriminative training of nai̇ve bayes classifiers for natural language call routing /." 2004.

Знайти повний текст джерела

Анотація:

Thesis (M.Sc.)--York University, 2004. Graduate Programme in Computer Science and Engineering. Typescript. Includes bibliographical references (leaves 76-79). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://gateway.proquest.com/openurl?url%5Fver=Z39.88-2004&res%5Fdat=xri:pqdiss &rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:MR11842

Стилі APA, Harvard, Vancouver, ISO та ін.

20

(8715537), Kyuseo Han. "Articulated Human Movements Tracking Through Online Discriminative Learning." Thesis, 2020.

Знайти повний текст джерела

Анотація:

In this thesis, we present a new class of object trackers that are based ona boosted Multiple Instance Learning (MIL) algorithm to track an object in a video sequence. We show how the scope of such trackers can be expanded to the tracking of articulated movements by humans that frequently result in large frame-to-frame variations in the appearance of what needs to be tracked. To deal with the problems caused by such variations, we present a component-based MIL (CMIL) algorithm with boosted learning. The components are the output of an image segmentation algorithm and give the boosted MIL the additional degrees of freedom that it needs in order to deal with the large frame-to-frame variations associated with articulated movements. Furthermore we explored two enhancements of the basic CMIL tracking algorithm. The first is based on an extended definition of positive learning samples for CMIL tracking. This extended definition can filter out false-positive learning samples in order to increase the robustness of CMIL tracking. The second enhancement is based on a combined motion prediction framework with the basic CMIL tracking for resolving issues arising from large and rapid translational human movements. The need for appropriate motion transition can be satisfied by probabilistic modeling of motion. Experimental results show that the proposed approaches yield robust tracking performances in various tracking environments, such as articulate human movements as well as ground human movements observed from aerial vehicles.

Стилі APA, Harvard, Vancouver, ISO та ін.

21

Asper, Roman Yorick. "Classifiers for Discrimination of Significant Protein Residues and Protein-Protein Interaction Using Concepts of Information Theory and Machine Learning." Doctoral thesis, 2011. http://hdl.handle.net/11858/00-1735-0000-0006-B3F3-E.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

22

MISHRA, Anamika. "Species Discrimination and Monitoring of Abiotic Stress Tolerance by Chlorophyll Fluorescence Transients." Doctoral thesis, 2012. http://www.nusl.cz/ntk/nusl-55608.

Повний текст джерела

Анотація:

Chlorophyll fluorescence imaging has now become a versatile and standard tool in fundamental and applied plant research. This method captures time series images of the chlorophyll fluorescence emission of whole leaves or plants upon various illuminations, typically combination of actinic light and saturating flashes. Several conventional chlorophyll fluorescence parameters have been recognized that have physiological interpretation and are useful for, e.g., assessment of plant health status and early detection of biotic and abiotic stresses. Chlorophyll florescence imaging enabled us to probe the performance of plants by visualizing physiologically relevant fluorescence parameters reporting physiology and biochemistry of the plant leaves. Sometimes there is a need to find the most contrasting fluorescence features/parameters in order to quantify stress response at very early stage of the stress treatment. The conventional fluorescence utilizes well defined single image such as F0, Fp, Fm, Fs or arithmetic combinations of basic images such as Fv/Fm, PSII, NPQ, qP. Therefore, although conventional fluorescence parameters have physiological interpretation, they may not be representing highly contrasting image sets. In order to find the effect of stress treatments at very early stage, advanced statistical techniques, based on classifiers and feature selection methods, have been developed to select highly contrasting chlorophyll fluorescence images out of hundreds of captured images. We combined sets of highly performing images resulting in images with very high contrast, the so called combinatorial imaging. The application of advanced statistical methods on chlorophyll fluorescence imaging data allows us to succeed in tasks, where conventional approaches do not work. This thesis aims to explore the application of conventional chlorophyll fluorescence parameters as well as advanced statistical techniques of classifiers and feature selection methods for high-throughput screening. We demonstrate the applicability of the technique in discriminating three species of the same family Lamiaceae at a very early stage of their growth. Further, we show that chlorophyll fluorescence imaging can be used for measuring cold and drought tolerance of Arabidopsis thaliana and tomato plants, respectively, in a simulated high ? throughput screening.

Стилі APA, Harvard, Vancouver, ISO та ін.

23

(9779444), Nathan Barrett. "Rainbow suits at work: Disclosure and discrimination in the workplace against gay, lesbian, bisexual, transgender and intersex employees." Thesis, 2011. https://figshare.com/articles/thesis/Rainbow_suits_at_work_Disclosure_and_discrimination_in_the_workplace_against_gay_lesbian_bisexual_transgender_and_intersex_employees/13457786.

Повний текст джерела

Анотація:

"This exploratory study examined people who self identify as gay, lesbian, bisexual, transgender or intersex (GLBTI) to determine the extent of workplace discrimination based on sexual identity in Queensland" -- Abstract.

Стилі APA, Harvard, Vancouver, ISO та ін.

24

(10746663), Samantha A. Peachey. "Examining Sexual and Relationship Satisfaction as Influenced by the Connection Between Sex Positivity and Perceived Discrimination for Sexual Minority Couples." Thesis, 2021.

Знайти повний текст джерела

Анотація:

The purpose of this research study was to look at the effects of perceived discrimination and sexual positivity on relationship and sexual satisfaction of sexual minority couples. The present study hypothesizes that there will be a moderating relationship between sexual positivity and perceived discrimination; higher levels of sexual positivity will predict higher relationship and sexual satisfaction, and perceived discrimination will negatively effect relationship and sexual satisfaction of couples with lower sexual positivity. Individuals who identify as a sexual minority were asked to participate in this study and answer survey questions pertaining to the level of satisfaction they experience in their romantic relationship and their sexual relationship, how sex positive the individuals are, and the amount of perceived discrimination that they experience; all through a minority stress lens. The results suggest that neither perceived discrimination nor the interaction between perceived discrimination and sexual positivity has a significant impact on the relationship and sexual satisfaction of sexual minority populations. However, the results of this study do suggest a statistically significant relationship between sexual positivity and relationship and sexual satisfaction of sexual minority couples.

Стилі APA, Harvard, Vancouver, ISO та ін.

25

Pham, Tung Huy. "Some problems in high dimensional data analysis." 2010. http://repository.unimelb.edu.au/10187/8399.

Повний текст джерела

Анотація:

The bloom of economics and technology has had an enormous impact on society. Along with these developments, human activities nowadays produce massive amounts of data that can be easily collected for relatively low cost with the aid of new technologies. Many examples can be mentioned here including data from web term-document data, sensor arrays, gene expression, finance data, imaging and hyperspectral analysis. Because of the enormous amount of data from various different and new sources, more and more challenging scientific problems appear. These problems have changed the types of problems which mathematical scientists work. In traditional statistics, the dimension of the data, p say, is low, with many observations, n say. In this case, classical rules such as the Central Limit Theorem are often applied to obtain some understanding from data. A new challenge to statisticians today is dealing with a different setting, when the data dimension is very large and the number of observations is small. The mathematical assumption now could be p > n, or even p goes to infinity and n fixed in many cases, for example, there are few patients with many genes. In these cases, classical methods fail to produce a good understanding of the nature of the problem. Hence, new methods need to be found to solve these problems. Mathematical explanations are also needed to generalize these cases. The research preferred in this thesis includes two problems: Variable selection and Classification, in the case where the dimension is very large. The work on variable selection problems, in particular the Adaptive Lasso was completed by June 2007 and the research on classification has been carried out through out 2008 and 2009. The research on the Dantzig selector and the Lasso were finished in July 2009. Therefore, this thesis is divided into two parts. In the first part of the thesis we study the Adaptive Lasso, the Lasso and the Dantzig selector. In particular, in Chapter 2 we present some results for the Adaptive Lasso. Chapter 3 will provides two examples that show that neither the Dantzig selector or the Lasso is definitely better than the other. The second part of the thesis is organized as follows. In Chapter 5, we shall construct the model setting. In Chapter 6, we summarize the results of the scaled centroid-based classifier. We also prove some results on the scaled centroid-based classifier. Because there are similarities between the Support Vector Machine (SVM) and Distance Weighted Discrimination (DWD) classifiers, Chapter 8 introduces a class of distance-based classifiers that could be considered a generalization of the SVM and DWD classifiers. Chapters 9 and 10 are about the SVM and DWD classifiers. Chapter 11 demonstrates the performance of these classifiers on simulated data sets and some cancer data sets.

Стилі APA, Harvard, Vancouver, ISO та ін.

26

(9160868), Jinho Jung. "ESSAYS ON SPATIAL DIFFERENTIATION AND IMPERFECT COMPETITION IN AGRICULTURAL PROCUREMENT MARKETS." Thesis, 2020.

Знайти повний текст джерела

Анотація:

<div> First Essay: We study the effect of entry of ethanol plants on the spatial pattern of corn prices. We use pre- and post-entry data from corn elevators to implement a clean identification strategy that allows us to quantify how price effects vary with the size of the entrant (relative to local corn production) and with distance from the elevator to the entrant. We estimate Difference-In-Difference (DID) and DID-matching models with linear and non-linear distance specifications. We find that the average-sized entrant causes an increase in corn price that ranges from 10 to 15 cents per bushel at the plant’s location, depending on the model specification. We also find that, on average, the price effect dissipates 60 miles away from the plant. Our results indicate that the magnitude of the price effect as well as its spatial pattern vary substantially with the size of the entrant relative to local corn supply. Under our preferred model, the largest entrant in our sample causes an estimated price increase of 15 cents per bushel at the plant’s site and the price effect propagates over 100 miles away. In contrast, the smallest entrant causes a price increase of only 2 cents per bushel at the plant’s site and the price effect dissipates within 15 miles of the plant. Our results are qualitatively robust to the pre-treatment matching strategy, to whether spatial effects are assumed to be linear or nonlinear, and to placebo tests that falsify alternative explanations. </div> Second Essay: We estimate the cost of transporting corn and the resulting degree of spatial differentiation among downstream firms that buy corn from upstream farmers and examine whether such differentiation softens competition enabling buyers to exert market power (defined as the ability to pay a price for corn that is below its marginal value product net of processing cost). We estimate a structural model of spatial competition using corn procurement data from the US state of Indiana from 2004 to 2014. We adopt a strategy that allows us to estimate firm-level structural parameters while using aggregate data. Our results return a transportation cost of 0.12 cents per bushel per mile (3% of the corn price under average conditions), which provides evidence of spatial differentiation among buyers. The estimated average markdown is $0.80 per bushel (16% of the average corn price in the sample), of which $0.34 is explained by spatial differentiation and the rest by the fact that firms operated under binding capacity constraints. We also find that corn prices paid to farmers at the mill gate are independent of distance between the plant and the farm, providing evidence that firms do not engage in spatial price discrimination. Finally, we evaluate the effect of hypothetical mergers on input markets and farm surplus. A merger between nearby ethanol producers eases competition, increases markdowns by 20%, and triggers a sizable reduction in farm surplus. In contrast, a merger between distant buyers has little effect on competition and markdowns. Third Essay: We study the dynamic response of local corn prices to entry of ethanol plants. We use spatially explicit panel data on elevator-level corn prices and ethanol plant entry and capacity to estimate an autoregressive distributed lag model with instrumental variables. We find that the average-sized entrant has no impact on local corn prices the year of entry. However, the price subsequently rises and stabilizes after two years at a level that is about 10 cents per bushel higher than the pre-entry level. This price effect dissipates as the distance between elevators and plants increase. Our results imply that long-run (2 years) supply elasticity is smaller than short-run (year of entry) supply elasticity. This may be due to rotation benefits that induce farmers to revert back to soybeans, after switching to corn due to price signals the year the plant enters. Furthermore, our results, in combination with findings in essay 2 of this dissertation, indicate that ethanol plants are likely to use pricing strategies consistent with a static rather than dynamic oligopsony competition.

Стилі APA, Harvard, Vancouver, ISO та ін.

27

(9790610), Carolyn Daniels. "Exploring Australian women's career transitions: A critical constructivist grounded theory study." Thesis, 2019. https://figshare.com/articles/thesis/Exploring_Australian_women_s_career_transitions_A_critical_constructivist_grounded_theory_study/13453955.

Повний текст джерела

Анотація:

Multiple factors impact women’s lives after they have transitioned through higher education and into the workforce, including the struggle to achieve a balance between paid work and unpaid care responsibilities. Despite changes in Australian social and cultural attitudes, career transitions remain difficult for many women in Australia. This thesis presents a qualitative, grounded theory study that explores how women in Australia navigate and experience career transitions; specifically, the transitions to higher education and the workforce which form the pathway to their careers. Emerging from the interviews were two distinct groups of women. Group 1 comprised mature age students on entry to university and Group 2 women had transitioned directly from high school to university. The Group 1 women shared stories about the lack of family and partner support, of time and sleep deprivation, and for some, domestic violence, dominance, resistance and abandonment as they studied. Many of these women experienced an up to fivefold burden of time as they juggled work, care, study, domestic responsibilities, and small business management. Concurrently, many experienced ‘pressure’, ‘guilt’, ‘stress’, changed and broken relationships and financial insecurity. Despite this, higher education held the promise of future security. Group 2 women invariably knew they would go to university, and were supported by their families. Those who married and later undertook further studies were supported by their husbands. For both groups, the university transition experience was influenced by varying degrees of support from families, partners, universities, social networks and relevant higher education and government policies. In the work transition experience however, it was the individual level of resilience required relative to available support that was the significant factor. Initially, a constructivist grounded theory approach was adopted. However, emergent social justice issues arising during interviews and analysis prompted development of a new methodology incorporating a critical perspective. Exploring the integration of a critical perspective required reviewing and juxtaposing constructivist grounded theory with critical theory. The shared axiomatic elements of these paradigms made them commensurable, and adaption made possible the concurrent practice of both. Thus, merging critical theory with constructivist grounded theory resulted in the new methodology critical constructivist grounded theory. It follows the evolutionary path of qualitative interpretive work, addressing the need for a critical stance to expose social justice issues. At the heart of the new methodology is a systematic analysis method, critical colours, enabling examination of the social, political, cultural, economic, structural, gender and historical forces impacting Australian women’s lives. What is more, critical colours analysis processes are adaptable to other axiologically congruent methodologies. Advanced coding methods identified the categories of Time Related Forces, Striving for Security, Transformation of Self and the construct of the Emergent Core Self. Theoretical integration of these categories and construct with critical colours analyses produced the critical constructivist grounded theory of Australian women’s career transitions. The grounded theory reveals the ideologies of neoliberalism, capitalism and patriarchy impose a restrictive framework to the ways in which women’s career transitions are experienced. The model illustrates that the more roles and responsibilities women assume, the greater the time and financial deficits they experience, the more support is required. The grounded theory crystallises women’s experiences as the cumulative effect of the forces of time, study, deprivations, the quest for security (financial and emotional), and the transformative power of learning. The Emergent Core Self makes clear that the women have an altered sense of resilience and knowledge of ‘who I am’. This study reveals that the navigation of career transitions by women in Australia is influenced by the undercurrents of an authoritarian social system skewed to inequality. The implications of the findings point strongly to the need for systemic change where equality is a matter-of-course. From a social systems perspective, it is suggested that the Nordic model provides a system of governance that benefits all citizens. This alternative model offers a solution to embed equity into the Australian systems of governance and social supports. The citizen-centred support characteristics of the Nordic model not only releases women’s burden of time and security, potentially delivering the supportive environment necessary for women in Australian to successfully navigate career transitions, it also provides a platform for equity for all.

Стилі APA, Harvard, Vancouver, ISO та ін.

Дисертації з теми "Discriminative classifier"

Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями