The mainly tropical genus Lygodium differs from other ferns in that the fronds are indeterminate and are vine-like. A single species, L. palmatum is native in temperate North America. The temperate Asian L. japonicum is naturalized throughout much of the southeastern United States. About twenty years ago, L. microphyllum was introduced into South Florida and is now naturalized in several counties. The present study documents differences among spores and their generation, development of sporophytes from the fertilized egg, and in flavonoid chemistry. Hybridization experiments showed a strong possibility for cross fertility between species. Experiments with prothallial development and differentiation revealed that environment influenced variation and gametangium formation. Greater similarity in sporophyte developmental stages and in frond phytochemistry show that the native L. palmatum is phenetically closer to the tropical L. microphyllum than to L. japonicum. All three species are clearly distinct at all levels examined.

Section I: Facultative and obligatory anaerobic bacteria were isolated from the ground water of a sanitary landfill characterized by sandy soil and a periodically high water table. Isolates were examined for 63 characteristics and subject to numerical analysis. Eight clusters were established and correlations with conventional taxonomy were made. The Bacteriodaceae were found to be the dominant group of organisms by the methods employed. The anaerobic population was observed to decrease as the period of seasonal rainfall ended. At the same time, gram positive anaerobes were largely replaced with gram negative ones. Leaching between sampling sites (wells) made correlations between metabolic end products (observed by gas-liquid chromatography) and metabolites produced by the organisms in vitro, impossible. Attempts were made to modify the original test battery to create a smaller battery which would yield approximately the same groupings as the original battery. Clusters became less discreet with these modifications and probably unacceptable for detailed taxonomic work. Section II: An index is described which measure the "goodness of fit" of an organism within a phenon as established by numerical taxonomy. A hypothetical mean organism was established for each phenon. Similarity and relevance coefficients were generated between this hypothetical organism and each member of the phenon. The product of these two coefficients has been termed the Index of Relevance and Similarity (IRS). This index ranges from zero to unity and can be generated with two-state and/or multistate data.

No / A robust method, fuzzy kNNModel, for toxicity prediction of chemical compounds is proposed. The method is based on a supervised clustering method, called kNNModel, which employs fuzzy partitioning instead of crisp partitioning to group clusters. The merits of fuzzy kNNModel are two-fold: (1) it overcomes the problems of choosing the parameter ¿ ¿ allowed error rate in a cluster and the parameter N ¿ minimal number of instances covered by a cluster, for each data set; (2) it better captures the characteristics of boundary data by assigning them with different degrees of membership between 0 and 1 to different clusters. The experimental results of fuzzy kNNModel conducted on thirteen public data sets from UCI machine learning repository and seven toxicity data sets from real-world applications, are compared with the results of fuzzy c-means clustering, k-means clustering, kNN, fuzzy kNN, and kNNModel in terms of classification performance. This application shows that fuzzy kNNModel is a promising method for the toxicity prediction of chemical compounds.

Automated defect detection methods allow the forest products industry to better utilize its resources by improving yield, reducing labor costs, and allowing minimum lumber grades to be utilized more intelligently. While many methods have been proposed on what sensors and processing methods should be used to detect and classify wood features, there exists a lack of understanding of what parameters are best used to differentiate wood features. The goal of this research is to demonstrate that by having an in depth knowledge of how wood features are represented by color, shape, and density parameters, more accurate classification methods can be developed. This goal was achieved through describing wood features using parameters derived from color and x-ray images and characterizing the variability and interrelationships of these parameters, determining the effect of resolution and species on these relationships, and determining the importance and contribution of each parameter for differentiating between wood features using a statistical prediction model relating feature types to the parameters. Knots, bark pockets, stain and mineral streak, and clearwood were selected as features from red oak, (Quercus rubra), hard maple, (Acer saccharum), and Eastern white pine (Pinus stobus). Color (RGB and HSI), shape (eccentricity and roundness), and density (gray-scale values) parameters were measured. Parameters were measured for each wood feature from images and parameter differences between feature types were tested using analysis of variance techniques (ANOVA) and Tukey's pairwise comparisons with a=0.05. Discriminant classifiers were then developed to demonstrate that an in-depth knowledge of how parameters relate between feature types could be used to develop the best possible classification methods. Classifiers developed using the knowledge of parameter relationships were found to provide higher classification accuracies for all features and species than those which used all parameters and where variable selection procedures had been used< It was determined that differences exist between all feature types and can be characterized and classified based on two color means, one color standard deviation, the mean density, and a shape parameter. A reduction in image resolution was determined not to affect the relationship of parameters. For different species, the intensity of features was to be related to the intensity of clearwood. The ability to explain classification errors using the knowledge gained about feature parameters was demonstrated. This knowledge could be used to reduce future classification errors. It was determined that combining parameters collected using multiple sensors increases classification accuracy of wood features. Shape and density were found not to provide good classification variables for features when used separately, but were found to contribute to classification of features when used with other parameters. The ability to differentiate between the feature types examined in this research was found be equal when using the RGB or HSI colorspace. / Ph. D.

Le but de cette étude est d'évaluer la stabilité temporelle d'un système de classification à trois profils conçu pour offrir une évaluation systématique de l'adaptation psychosociale au diabète (MCP-D). Le MCP-D regroupe les patients diabétiques sur la base des variables cognitives et sociales mesurées par le Questionnaire d'Évaluation Multidimensionnelle du Diabète (QMD). Deux prises de mesures séparées par un intervalle de neuf mois permettent de classifier à deux reprises 128 sujets diabétiques de type II. L'indice Kappa du degré de concordance entre les deux matrices de classification est de . 70, ce qui appuie adéquatement la stabilité temporelle des profils de caractéristiques psychosociales. Pour les patients qui ont changé de profils, les résultats révèlent des changements significatifs sur les variables psychologiques générales. Par ailleurs, ces derniers patients ne présentent aucune variation significative au niveau des variables psychologiques de motivation et des variables physiologiques reliées à la maladie. Ces résultats sont interprétés en fonction de la durée de l'intervalle de temps séparant les deux prises de mesures.

Classification of tissues and diseases based on gene expression data is a powerful application of DNA microarrays. Many popular classifiers like support vector machines, nearest-neighbour methods, and boosting have been applied successfully to this problem. However, it is difficult to determine from these classifiers which genes are responsible for the distinctions between the diseases. We propose a novel framework for classification of gene expression data based on notion of condition-specific clusters of co-expressed genes called xMotifs. Our xMotif-based classifier is biologically interpretable: we show how we can detect relationships between xMotifs and gene functional annotations. Our classifier achieves high-accuracy on leave-one-out cross-validation on both two-class and multi-class data. Our technique has the potential to be the method of choice for researchers interested in disease and tissue classification. / Master of Science

This thesis describes the ideal properties of an adaptable music classification system based on unsupervised machine learning, and argues that such a system should be based on the fundamental musical properties of timbre, rhythm, melody and harmony. The first two properties and the signal features associated with them are then explored in more depth. In the area of timbre, the relationship between musical style and commonly-extracted signal features within a broad range of piano music is explored, in an effort to identify features which are consistent among all piano music but different for other instruments. The effect of lossy compression on these same timbre features is also investigated. In the area of rhythm, a new tempo tracking tool is provided which produces a series of histograms containing beat and sub-beat information throughout the course of a musical recording. These histograms are then shown to be useful in the analysis of synthesized rhythms and real music. Additionally, a novel method based on the Expectation-Maximization algorithm is used to extract features for classification from the histograms. / Master of Science

This study examined the racial identification of biracial Asian children by their parents, in a sample (N=9,513) drawn from 2000 Public Use Microdata Series Census data (Integrated Public Use Microdata Series 2009). I used competing theories of Asian assimilation to examine how characteristics of the child, the Asian parent, the non-Asian parent, and the local Asian community influenced the likelihood of a child's being identified as Asian, non-Asian, or biracial. Findings showed that child's, both parents', and community characteristics significantly influenced the child's racial classification. While the effects of greater assimilation significantly increased the likelihood of an Asian classification for third-generation children, in contrast, it decreased the likelihood of an Asian identification for first- and second-generation children. Findings showed that children with a black parent were less likely than children with a white parent to be identified as Asian instead of non-Asian. However, inconsistent with past findings, children with a Hispanic parent were more likely than those with a white parent to be identified as Asian rather than non-Asian. Exploratory analyses concerning a biracial classification indicate significant relationships with factors previously found to increase the likelihood of an Asian identification, including the effects of greater Asian assimilation and size of the local Asian community. Moreover, the relationship between parent's and child's gender on the child's racial classification may be more complicated than previously theorized, as I found evidence of "gender-matching" which meant that boys were more likely to be identified like their fathers, and girls more like their mothers. / Master of Science

Gas Chromatography (GC) is a technique for separating volatile compounds by relying on adherence differences in the chemical components of the compound. As conditions within the GC are changed, components of the mixture elute at different times. Sensors measure the elution and produce data which becomes chromatograms. By analyzing the chromatogram, the presence and quantity of the mixture's constituent components can be determined. Machine Learning (ML) is a field consisting of techniques by which machines can independently analyze data to derive their own procedures for processing it. Additionally, there are techniques for enhancing the performance of ML algorithms. Feature Selection is a technique for improving performance by using a specific subset of the data. Feature Engineering is a technique to transform the data to make processing more effective. Data Fusion is a technique which combines multiple sources of data so as to produce more useful data. This thesis applies machine learning algorithms to chromatograms. Five common machine learning algorithms are analyzed and compared, including K-Nearest Neighbour (KNN), Support Vector Machines (SVM), Convolutional Neural Network (CNN), Decision Tree, and Random Forest (RF). Feature Selection is tested by applying window sweeps with the KNN algorithm. Feature Engineering is applied via the Principal Component Analysis (PCA) algorithm. Data Fusion is also tested. It was found that KNN and RF performed best overall. Feature Selection was very beneficial overall. PCA was helpful for some algorithms, but less so for others. Data Fusion was moderately beneficial. / Master of Science / Gas Chromatography is a method for separating a mixture into its constituent components. A chromatogram is a time series showing the detection of gas in the gas chromatography machine over time. With a properly set up gas chromatographer, different mixtures will produce different chromatograms. These differences allow researchers to determine the components or differentiate compounds from each other. Machine Learning (ML) is a field encompassing a set of methods by which machines can independently analyze data to derive the exact algorithms for processing it. There are many different machine learning algorithms which can accomplish this. There are also techniques which can process the data to make it more effective for use with machine learning. Feature Engineering is one such technique which transforms the data. Feature Selection is another technique which reduces the data to a subset. Data Fusion is a technique which combines different sources of data. Each of these processing techniques have many different implementations. This thesis applies machine learning to gas chromatography. ML systems are developed to classify mixtures based on their chromatograms. Five common machine learning algorithms are developed and compared. Some common Feature Engineering, Feature Selection, and Data Fusion techniques are also evaluated. Two of the algorithms were found to be more effective overall than the other algorithms. Feature Selection was found to be very beneficial. Feature Engineering was beneficial for some algorithms but less so for others. Data Fusion was moderately beneficial.

Yes / Contrast discrimination measures the smallest difference in contrast (the threshold) needed to successfully tell two stimuli apart. The contrast discrimination threshold typically increases with contrast. However, for low spatial frequency gratings the contrast threshold first increases, but then starts to decrease at contrasts above about 50%. This behaviour was originally observed in contrast discrimination experiments using dark spots as stimuli, suggesting that the contrast discrimination threshold for low spatial frequency gratings may be dominated by responses to the dark parts of the sinusoid. This study measures classification images for contrast discrimination experiments using a 1 cycle per degree sinusoidal grating at contrasts of 0, 25%, 50% and 75%. The classification images obtained clearly show that observers emphasize the darker parts of the sinusoidal grating (i.e. the troughs), and this emphasis increases with contrast. At 75% contrast, observers almost completely ignored the bright parts (peaks) of the sinusoid, and for some observers the emphasis on the troughs is already evident at contrasts as low as 25%. Analysis using a Hammerstein model suggests that the bias towards the dark parts of the stimulus is due to an early nonlinearity, perhaps similar to that proposed by Whittle.