Novel Regression Methods For Spectral Data

Singular Value Decomposition (SVD), Principal Component Analysis (PCA) and Multiple Linear Regression (MLR) are some of the mathematical pre- liminaries that are discussed prior to explaining PLS and PCR models. Both PLS and PCR are applied to real spectral data and their di erences and similarities are discussed in this thesis. The challenge lies in establishing the optimum number of components to be included in either of the models but this has been overcome by using various diagnostic tools suggested in this thesis. Correspondence analysis (CA) and PLS were applied to ecological data. The idea of CA was to correlate the macrophytes species and lakes. The di erences between PLS model for ecological data and PLS for spectral data are noted and explained in this thesis. i

Efficient Optimization Algorithms for Nonlinear Data Analysis

Identification of low-dimensional structures and main sources of variation from multivariate data are fundamental tasks in data analysis. Many methods aimed at these tasks involve solution of an optimization problem. Thus, the objective of this thesis is to develop computationally efficient and theoretically justified methods for solving such problems. Most of the thesis is based on a statistical model, where ridges of the density estimated from the data are considered as relevant features. Finding ridges, that are generalized maxima, necessitates development of advanced optimization methods. An efficient and convergent trust region Newton method for projecting a point onto a ridge of the underlying density is developed for this purpose. The method is utilized in a differential equation-based approach for tracing ridges and computing projection coordinates along them. The density estimation is done nonparametrically by using Gaussian kernels. This allows application of ridge-based methods with only mild assumptions on the underlying structure of the data. The statistical model and the ridge finding methods are adapted to two different applications. The first one is extraction of curvilinear structures from noisy data mixed with background clutter. The second one is a novel nonlinear generalization of principal component analysis (PCA) and its extension to time series data. The methods have a wide range of potential applications, where most of the earlier approaches are inadequate. Examples include identification of faults from seismic data and identification of filaments from cosmological data. Applicability of the nonlinear PCA to climate analysis and reconstruction of periodic patterns from noisy time series data are also demonstrated. Other contributions of the thesis include development of an efficient semidefinite optimization method for embedding graphs into the Euclidean space. The method produces structure-preserving embeddings that maximize interpoint distances. It is primarily developed for dimensionality reduction, but has also potential applications in graph theory and various areas of physics, chemistry and engineering. Asymptotic behaviour of ridges and maxima of Gaussian kernel densities is also investigated when the kernel bandwidth approaches infinity. The results are applied to the nonlinear PCA and to finding significant maxima of such densities, which is a typical problem in visual object tracking.

Dimensionality of the premenstrual syndrome: confirmatory factor analysis of premenstrual dysphoric symptoms among college students

Premenstrual syndrome and premenstrual dysphoric disorder (PMDD) seem to form a severity continuum with no clear-cut boundary. However, since the American Psychiatric Association proposed the research criteria for PMDD in 1994, there has been no agreement about the symptomatic constellation that constitutes this syndrome. The objective of the present study was to establish the core latent structure of PMDD symptoms in a non-clinical sample. Data concerning PMDD symptoms were obtained from 632 regularly menstruating college students (mean age 24.4 years, SD 5.9, range 17 to 49). For the first random half (N = 316), we performed principal component analysis (PCA) and for the remaining half (N = 316), we tested three theory-derived competing models of PMDD by confirmatory factor analysis. PCA allowed us to extract two correlated factors, i.e., dysphoric-somatic and behavioral-impairment factors. The two-dimensional latent model derived from PCA showed the best overall fit among three models tested by confirmatory factor analysis (c²53 = 64.39, P = 0.13; goodness-of-fit indices = 0.96; adjusted goodness-of-fit indices = 0.95; root mean square residual = 0.05; root mean square error of approximation = 0.03; 90%CI = 0.00 to 0.05; Akaike's information criterion = -41.61). The items "out of control" and "physical symptoms" loaded conspicuously on the first factor and "interpersonal impairment" loaded higher on the second factor. The construct validity for PMDD was accounted for by two highly correlated dimensions. These results support the argument for focusing on the core psychopathological dimension of PMDD in future studies.

Long-term social recognition memory in adult male rats: factor analysis of the social and non-social behaviors

A modified version of the intruder-resident paradigm was used to investigate if social recognition memory lasts at least 24 h. One hundred and forty-six adult male Wistar rats were used. Independent groups of rats were exposed to an intruder for 0.083, 0.5, 2, 24, or 168 h and tested 24 h after the first encounter with the familiar or a different conspecific. Factor analysis was employed to identify associations between behaviors and treatments. Resident rats exhibited a 24-h social recognition memory, as indicated by a 3- to 5-fold decrease in social behaviors in the second encounter with the same conspecific compared to those observed for a different conspecific, when the duration of the first encounter was 2 h or longer. It was possible to distinguish between two different categories of social behaviors and their expression depended on the duration of the first encounter. Sniffing the anogenital area (49.9% of the social behaviors), sniffing the body (17.9%), sniffing the head (3%), and following the conspecific (3.1%), exhibited mostly by resident rats, characterized social investigation and revealed long-term social recognition memory. However, dominance (23.8%) and mild aggression (2.3%), exhibited by both resident and intruders, characterized social agonistic behaviors and were not affected by memory. Differently, sniffing the environment (76.8% of the non-social behaviors) and rearing (14.3%), both exhibited mostly by adult intruder rats, characterized non-social behaviors. Together, these results show that social recognition memory in rats may last at least 24 h after a 2-h or longer exposure to the conspecific.

Correlation, by multivariate statistical analysis, between the scavenging capacity against reactive oxygen species and the bioactive compounds from frozen fruit pulps

The contents of total phenolic compounds (TPC), total flavonoids (TF), and ascorbic acid (AA) of 18 frozen fruit pulps and their scavenging capacities against peroxyl radical (ROO•), hydrogen peroxide (H2O2), and hydroxyl radical (•OH) were determined. Principal Component Analysis (PCA) showed that TPC (total phenolic compounds) and AA (ascorbic acid) presented positive correlation with the scavenging capacity against ROO•, and TF (total flavonoids) showed positive correlation with the scavenging capacity against •OH and ROO• However, the scavenging capacity against H2O2 presented low correlation with TF (total flavonoids), TPC (total phenolic compounds), and AA (ascorbic acid). The Hierarchical Cluster Analysis (HCA) allowed the classification of the fruit pulps into three groups: one group was formed by the açai pulp with high TF, total flavonoids, content (134.02 mg CE/100 g pulp) and the highest scavenging capacity against ROO•, •OH and H2O2; the second group was formed by the acerola pulp with high TPC, total phenolic compounds, (658.40 mg GAE/100 g pulp) and AA , ascorbic acid, (506.27 mg/100 g pulp) contents; and the third group was formed by pineapple, cacao, caja, cashew-apple, coconut, cupuaçu, guava, orange, lemon, mango, passion fruit, watermelon, pitanga, tamarind, tangerine, and umbu pulps, which could not be separated considering only the contents of bioactive compounds and the scavenging properties.

Characterization and classification of pequi trees (Caryocar brasiliense Camb.) based on the profile of volatile constituents using headspace solid-phase microextraction - gas chromatography - mass spectrometry and multivariate analysis

In order to determine the variability of pequi tree (Caryocar brasiliense Camb.) populations, volatile compounds from fruits of eighteen trees representing five populations were extracted by headspace solid-phase microextraction and analyzed by gas chromatography-mass spectrometry. Seventy-seven compounds were identified, including esters, hydrocarbons, terpenoids, ketones, lactones, and alcohols. Several compounds had not been previously reported in the pequi fruit. The amount of total volatile compounds and the individual compound contents varied between plants. The volatile profile enabled the differentiation of all of the eighteen plants, indicating that there is a characteristic profile in terms of their origin. The use of Principal Component Analysis and Cluster Analysis enabled the establishment of markers (dendrolasin, ethyl octanoate, ethyl 2-octenoate and β-cis-ocimene) that discriminated among the pequi trees. According to the Cluster Analysis, the plants were classified into three main clusters, and four other plants showed a tendency to isolation. The results from multivariate analysis did not always group plants from the same population together, indicating that there is greater variability within the populations than between pequi tree populations.

Multivariate study and regression analysis of gluten-free granola

This study developed a gluten-free granola and evaluated it during storage with the application of multivariate and regression analysis of the sensory and instrumental parameters. The physicochemical, sensory, and nutritional characteristics of a product containing quinoa, amaranth and linseed were evaluated. The crude protein and lipid contents ranged from 97.49 and 122.72 g kg-1 of food, respectively. The polyunsaturated/saturated, and n-6:n-3 fatty acid ratios ranged from 2.82 and 2.59:1, respectively. Granola had the best alpha-linolenic acid content, nutritional indices in the lipid fraction, and mineral content. There were good hygienic and sanitary conditions during storage; probably due to the low water activity of the formulation, which contributed to inhibit microbial growth. The sensory attributes ranged from 'like very much' to 'like slightly', and the regression models were highly fitted and correlated during the storage period. A reduction in the sensory attribute levels and in the product physical stabilisation was verified by principal component analysis. The use of the affective test acceptance and instrumental analysis combined with statistical methods allowed us to obtain promising results about the characteristics of gluten-free granola.

A comparative study among methods used for wheat flour analysis and for measurements of gluten properties using the Wheat Gluten Quality Analyser (WGQA)

This study aimed at comparing both the results of wheat flour quality assessed by the new equipment Wheat Gluten Quality Analyser (WGQA) and those obtained by the extensigraph and farinograph. Fifty-nine wheat samples were evaluated for protein and gluten contents; the rheological properties of gluten and wheat flour were assessed using the WGQA and the extensigraph/farinograph methods, respectively, in addition to the baking test. Principal component analysis (PCA) and linear regression were used to evaluate the results. The parameters of energy and maximum resistance to extension determined by the extensigraph and WGQA showed an acceptable level for the linear correlation within the range from 0.6071 to 0.6511. The PCA results obtained using WGQA and the other rheological apparatus showed values similar to those expected for wheat flours in the baking test. Although all equipment used was effective in assessing the behavior of strong and weak flours, the results of medium strength wheat flour varied. WGQA has shown to use less amount of sample and to be faster and easier to use in relation to the other instruments used.

Corporate Entrepreneurship in Healthcare Organizations: An explorative study of employees’ corporate entrepreneurial behavior, job satisfaction and turnover intention

The Finnish healthcare industry is currently facing significant challenges due to economic crises, aging population and major structural reforms, which have resulted in decreased job satisfaction and increased levels of turnover. This proposes that healthcare organizations need to come up with new, creative means to tackle these issues. Several researchers have argued that corporate entrepreneurship may be the necessary means to achieve this. As previous research has mainly focused on examining this concept from organizational perspective, this study looks at how it occurs on the level of individual employees. The purpose of this study is to examine how corporate entrepreneurship is manifested in individual behavior, and how this type of behavior is associated with the individual’s job satisfaction and turnover intention. Additionally, this study will examine the differences in corporate entrepreneurial behavior between private and public sector organizations, as previous research suggests that these two may be characterized differently. Data was collected with the help of a literature review as well as a survey study, which was sent out to a number of employees of four different healthcare organizations, out of which three were public and one was a private sector organization. Six distinct behavioral characteristics were recognized in previous research, which make up the measure for corporate entrepreneurial behavior. Principal components were formed from the different areas of the survey (corporate entrepreneurial behavior, job satisfaction, turnover intention), after which the association of these components were examined with linear regression analysis, which proved that corporate entrepreneurial behavior is positively correlated with both job satisfaction and intention to leave the organization. Differences between sectors were analyzed with analysis of variance and cross tabulation analysis, but neither of these suggested that any significant differences would occur. These results suggest that employees who behave entrepreneurially tend to be more satisfied with their jobs, but also consider leaving their current organizations more often than others. This may be due to the fact that healthcare organizations are not fertile for entrepreneurial behavior, which will drive entrepreneurial individuals looking for employers who may be more supportive of this type of behavior. With growing levels of dissatisfaction as well as little room for entrepreneurial behavior, the studied organizations may actually be in the process of losing those employees who have the ability and desire to behave in such manner, and who could very well be those who will eventually come up with solutions for the major challenges that these organizations are facing.

The Terroir of Winter Hardiness: Investigation of Winter Hardiness, Water Metrics, and Yield of Riesling and Cabernet Franc in the Niagara Region Using Geomatic Technologies

Grapevine winter hardiness is a key factor in vineyard success in many cool climate wine regions. Winter hardiness may be governed by a myriad of factors in addition to extreme weather conditions – e.g. soil factors (texture, chemical composition, moisture, drainage), vine water status, and yield– that are unique to each site. It was hypothesized that winter hardiness would be influenced by certain terroir factors , specifically that vines with low water status [more negative leaf water potential (leaf ψ)] would be more winter hardy than vines with high water status (more positive leaf ψ). Twelve different vineyard blocks (six each of Riesling and Cabernet franc) throughout the Niagara Region in Ontario, Canada were chosen. Data were collected during the growing season (soil moisture, leaf ψ), at harvest (yield components, berry composition), and during the winter (bud LT50, bud survival). Interpolation and mapping of the variables was completed using ArcGIS 10.1 (ESRI, Redlands, CA) and statistical analyses (Pearson’s correlation, principal component analysis, multilinear regression) were performed using XLSTAT. Clear spatial trends were observed in each vineyard for soil moisture, leaf ψ, yield components, berry composition, and LT50. Both leaf ψ and berry weight could predict the LT50 value, with strong positive correlations being observed between LT50 and leaf ψ values in eight of the 12 vineyard blocks. In addition, vineyards in different appellations showed many similarities (Niagara Lakeshore, Lincoln Lakeshore, Four Mile Creek, Beamsville Bench). These results suggest that there is a spatial component to winter injury, as with other aspects of terroir, in the Niagara region.

Hyperspectral Mineral Identification using SVM and SOM

Remote sensing techniques involving hyperspectral imagery have applications in a number of sciences that study some aspects of the surface of the planet. The analysis of hyperspectral images is complex because of the large amount of information involved and the noise within that data. Investigating images with regard to identify minerals, rocks, vegetation and other materials is an application of hyperspectral remote sensing in the earth sciences. This thesis evaluates the performance of two classification and clustering techniques on hyperspectral images for mineral identification. Support Vector Machines (SVM) and Self-Organizing Maps (SOM) are applied as classification and clustering techniques, respectively. Principal Component Analysis (PCA) is used to prepare the data to be analyzed. The purpose of using PCA is to reduce the amount of data that needs to be processed by identifying the most important components within the data. A well-studied dataset from Cuprite, Nevada and a dataset of more complex data from Baffin Island were used to assess the performance of these techniques. The main goal of this research study is to evaluate the advantage of training a classifier based on a small amount of data compared to an unsupervised method. Determining the effect of feature extraction on the accuracy of the clustering and classification method is another goal of this research. This thesis concludes that using PCA increases the learning accuracy, and especially so in classification. SVM classifies Cuprite data with a high precision and the SOM challenges SVM on datasets with high level of noise (like Baffin Island).

L'éthique en psychothérapie dans le cadre des processus de transfert et de contre-transfert: l'apport de la médecine narrative et du concept de représentation

Les discussions sur l’éthique de la santé mentale se concentrent généralement sur des sujets comme la confidentialité ou les conflits d’intérêt. Certaines formes de thérapie donnent cependant lieu à des situations plus particulières. C’est le cas de la thérapie psychodynamique et des considérations éthiques soulevées par les processus de transfert et de contre-transfert. En raison du peu de recherche sur ce sujet, l’approche de l’analyse documentaire critique, une méthode qualitative, fut sélectionnée. Ce projet vise à établir l’apport de la médecine narrative et plus spécifiquement du concept de représentation pour analyser les aspects éthiques du transfert et du contre-transfert. À cette fin, ces processus sont étudiés en détail et les diverses stratégies proposées aux thérapeutes pour maintenir une relation éthique avec leurs patients sont recensées. Popularisés principalement par Rita Charon, la médecine narrative et ses principaux composants, dont la représentation, sont ensuite présentés. De plus, la mimèsis étant à la base de la représentation, les travaux de Paul Ricœur sur le sujet sont discutés. Le dernier chapitre de ce mémoire, qui constitue la partie analytique du projet, prend la forme d’un exercice inductif évaluant l’apport de la médecine narrative pour analyser les aspects éthiques du transfert et du contre-transfert.

L’agence en oncologie de l’Ontario : actifs organisationnels et professionnels à l’amélioration des soins et des services

Cette thèse s’intéresse à l’amélioration des soins et des services de santé et touche aux relations entre 3 grands thèmes de l’analyse des organisations de santé : la gouvernance, le changement et les pratiques professionnelles. En nous appuyant sur l’analyse organisationnelle contemporaine, nous visons à mieux comprendre l’interface entre l’organisation et les pratiques cliniques. D’une part, nous souhaitons mieux comprendre comment l’organisation structure et potentialise les pratiques des acteurs. D’autre part, dans une perspective d’acteurs stratégiques, nous souhaitons mieux comprendre le rôle des pratiques des professionnels dans l’actualisation de leur profession et dans la transformation et l’évolution des organisations. Notre étude se fonde sur l’hypothèse qu’une synergie accrue entre l’organisation et les pratiques des professionnels favorisent l’amélioration de la qualité des soins et des services de santé. En 2004, le gouvernement ontarien entreprend une importante réforme des soins et services dans le domaine du cancer et revoit les rôles et mandats du Cancer Care Ontario, l’organisation responsable du développement des orientations stratégiques et du financement des services en cancer dans la province. Cette réforme appelle de nombreux changements organisationnels et cliniques et vise à améliorer la qualité des soins et des services dans le domaine de l’oncologie. C’est dans le cadre de cette réforme que nous avons analysé l’implantation d’un système de soins et de services pour améliorer la performance et la qualité et analysé le rôle des pratiques professionnelles, spécifiquement les pratiques infirmières, dans la transformation de ce système. La stratégie de recherche utilisée correspond à l’étude approfondie d’un cas correspondant à l’agence de soins et de services en oncologie en Ontario, le Cancer Care Ontario, et des pratiques professionnelles infirmières évoluant dans ce modèle. Le choix délibéré de ce cas repose sur les modalités organisationnelles spécifiques à l’Ontario en termes de soins en oncologie. La collecte de données repose sur 3 sources principales : les entrevues semi-structurées (n=25), l’analyse d’une abondante documentation et les observations non participatives. La thèse s’articule autour de trois articles. Le premier article vise à définir le concept de gouvernance clinique. Nous présentons l’origine du concept et définissons ses principales composantes. Concept aux frontières floues, la gouvernance clinique est axée sur le développement d’initiatives cliniques et organisationnelles visant à améliorer la qualité des soins de santé et la sécurité des patients. L’analyse de la littérature scientifique démontre la prédominance d’une vision statique de la gouvernance clinique et d’un contrôle accentué des pratiques professionnelles dans l’atteinte de l’efficience et de l’excellence dans les soins et les services. Notre article offre une conception plus dynamique de la gouvernance clinique qui tient compte de la synergie entre le contexte organisationnel et les pratiques des professionnels et soulève les enjeux reliés à son implantation. Le second article s’intéresse à l’ensemble des leviers mobilisés pour institutionnaliser les principes d’amélioration continue de la qualité dans les systèmes de santé. Nous avons analysé le rôle et la portée des leviers dans l’évolution du système de soins en oncologie en Ontario et dans la transformation des pratiques cliniques. Nos données empiriques révèlent 3 phases et de nombreuses étapes dans la transformation du système. Les acteurs en position d’autorité ont mobilisé un ensemble de leviers pour introduire des changements. Notre étude révèle que la transformation du Cancer Care Ontario est le reflet d’un changement radical de type évolutif où chacune des phases est une période charnière dans la transformation du système et l’implantation d’initiatives de qualité. Le troisième article pose un regard sur un levier spécifique de transformation, celui de la communauté de pratique, afin de mieux comprendre le rôle joué par les pratiques professionnelles dans la transformation de l’organisation des soins et ultimement dans le positionnement stratégique de la profession infirmière. Nous avons analysé les pratiques infirmières au sein de la communauté de pratique (CDP) des infirmières en pratique avancée en oncologie. En nous appuyant sur la théorie de la stratégie en tant que pratique sociale, nos résultats indiquent que l’investissement de la profession dans des domaines stratégiques augmente les capacités des infirmières à transformer leurs pratiques et à transformer l’organisation. Nos résultats soulignent le rôle déterminant du contexte dans le développement de capacités stratégiques chez les professionnels. Enfin, nos résultats révèlent 3 stratégies émergentes des pratiques des infirmières : une stratégie de développement de la pratique infirmière en oncologie, une stratégie d’institutionnalisation des politiques de la CDP dans le système en oncologie et une stratégie de positionnement de la profession infirmière. Les résultats de notre étude démontrent que l’amélioration de la qualité des soins et des services de santé est située. L’implantation de transformations dans l’ensemble d’un système, tel que celui du cancer en Ontario, est tributaire d’une part, des capacités d’action des acteurs en position d’autorité qui mobilisent un ensemble de leviers pour introduire des changements et d’autre part, de la capacité des acteurs à la base de l’organisation à s’approprier les leviers pour développer un projet professionnel, améliorer leurs pratiques professionnelles et transformer le système de soins.

Statistical methods for insurance fraud detection

Mémoire numérisé par la Division de la gestion de documents et des archives de l'Université de Montréal

Towards the identification of a neighbourhood park typology : a conceptual and methodological exploration

Peu d’études ont évalué les caractéristiques des parcs pouvant encourager l’activité physique spécifiquement chez les jeunes. Cette étude vise à estimer la fiabilité d’un outil d’observation des parcs orienté vers les jeunes, à identifier les domaines conceptuels des parcs capturés par cet outil à l’aide d’une opérationnalisation du modèle conceptuel des parcs et de l’activité physique et à identifier différents types de parcs. Un total de 576 parcs ont été évalués en utilisant un outil d’évaluation des parcs. La fiabilité intra-juges et la fiabilité inter-juges de cet outil ont été estimées. Une analyse exploratoire par composantes principales (ACP) a été effectuée en utilisant une rotation orthogonale varimax et les variables étaient retenues si elles saturaient à ≥0.3 sur une composante. Une analyse par grappes (AG) à l’aide de la méthode de Ward a ensuite été réalisée en utilisant les composantes principales et une mesure de l’aire des parcs. L’outil était généralement fiable et l’ACP a permis d'identifier dix composantes principales qui expliquaient 60% de la variance totale. L’AG a donné un résultat de neuf grappes qui expliquaient 40% de la variance totale. Les méthodes de l’ACP et l’AG sont donc faisables avec des données de parcs. Les résultats ont été interprétés en utilisant l’opérationnalisation du modèle conceptuel.