Exploring the integration of data mining and data visualization

Due to the rapid advances in computing and sensing technologies, enormous amounts of data are being generated everyday in various applications. The integration of data mining and data visualization has been widely used to analyze these massive and complex data sets to discover hidden patterns. For both data mining and visualization to be effective, it is important to include the visualization techniques in the mining process and to generate the discovered patterns for a more comprehensive visual view. In this dissertation, four related problems: dimensionality reduction for visualizing high dimensional datasets, visualization-based clustering evaluation, interactive document mining, and multiple clusterings exploration are studied to explore the integration of data mining and data visualization. In particular, we 1) propose an efficient feature selection method (reliefF + mRMR) for preprocessing high dimensional datasets; 2) present DClusterE to integrate cluster validation with user interaction and provide rich visualization tools for users to examine document clustering results from multiple perspectives; 3) design two interactive document summarization systems to involve users efforts and generate customized summaries from 2D sentence layouts; and 4) propose a new framework which organizes the different input clusterings into a hierarchical tree structure and allows for interactive exploration of multiple clustering solutions.

Tobacco use prevention education (TUPE) programs in the state of Florida: Teachers' perceptions of program structure and effectiveness

This study described teacher perceptions of TUPE program effectiveness in Florida in an attempt to improve programs by identifying factors that might influence teacher motivation and performance. Very little work has been done to examine how teachers' perceptions are related to the effectiveness of TUPE programs. A statewide survey provided information about how teachers' perceptions of program effectiveness are affected by variables such as: program structure, barriers, tobacco use norms, and training variables. Data were obtained from a telephone survey conducted in Florida as part of the Tobacco Pilot Project (TPP). The sample included 296 middle school teachers and 282 high school teachers as well as 193 middle school principals and 190 high school principals. Correlational and hierarchical regression analyses identified correlates and predictors of teachers' ratings of effectiveness. Results suggest that use of peer leaders, more frequent evaluations, a higher degree of parent involvement, fewer barriers, greater student interest, and lower tolerance for tobacco use were correlated with higher ratings of program effectiveness. Furthermore, student interest, peer, staff, and community tolerance norms, peer leaders, program evaluation, and parent involvement predicted middle school teachers' perceptions. Parent tolerance, student interest, number of barriers, and more frequent program evaluation predicted high school teachers' perceptions. In addition, middle school teachers who reported a lower number of factors negatively associated with teacher receptivity were more likely to view TUPE programs more favorably than teachers who reported a greater number of these risk factors. This relationship was not as robust among the high school teacher sample. Differences between the middle and high school sample were found in the magnitude and number of significant correlations, the proportion of variance accounted for by predictor variables, and the strength of the relationship between the number of factors negatively associated with teacher receptivity and teachers' perceptions of TUPE effectiveness. These findings highlighted the importance of the timing, program features, and the external environment for enhancing or minimizing teachers' ratings of TUPE program effectiveness. In conclusion, significant increases in TUPE teachers' self-efficacy will occur through the participation of peers, parents, staff, and community leaders in different aspects of TUPE programs. ^

Applying stable isotopes to examine food-web structure: an overview of analytical tools

Stable isotope analysis has emerged as one of the primary means for examining the structure and dynamics of food webs, and numerous analytical approaches are now commonly used in the field. Techniques range from simple, qualitative inferences based on the isotopic niche, to Bayesian mixing models that can be used to characterize food-web structure at multiple hierarchical levels. We provide a comprehensive review of these techniques, and thus a single reference source to help identify the most useful approaches to apply to a given data set. We structure the review around four general questions: (1) what is the trophic position of an organism in a food web?; (2) which resource pools support consumers?; (3) what additional information does relative position of consumers in isotopic space reveal about food-web structure?; and (4) what is the degree of trophic variability at the intrapopulation level? For each general question, we detail different approaches that have been applied, discussing the strengths and weaknesses of each. We conclude with a set of suggestions that transcend individual analytical approaches, and provide guidance for future applications in the field.

Enhancement of field emission from multistage structure of carbon nanotube arrays

The field emission measurements for the multistage structured nanotubes (i.e., thin-multiwall and single wall carbon nanotubes grown on multiwall carbon nanotubes) were carried out and a low turn-on field of ~0.45 V/ μm, high emission current of 450 μA at a field of IV/μm and a large field enhancement factor of ~26200 were obtained. The thin multiwall carbon nanotubes (thin-MWNTs) and single wall carbon nanotubes (SWNTs) were grown on the regular arrays of vertically aligned multi wall carbon nanotubes (MWNTs) on porous silicon substrate by Chemical Vapor Deposition (CVD) method. The thin-MWNTs and SWNTs grown on MWNTs in this way have a multistage structure which gives higher enhancement of the electric field and hence the electron field emission.

Meshfree Method for Prediction of Thermal Properties of Porous Ceramic Materials

In the presented thesis work, meshfree method with distance fields is applied to create a novel computational approach which enables inclusion of the realistic geometric models of the microstructure and liberates Finite Element Analysis(FEA) from thedependance on and limitations of meshing of fine microstructural feature such as splats and porosity.Manufacturing processes of ceramics produce materials with complex porosity microstructure.Geometry of pores, their size and location substantially affect macro scale physical properties of the material. Complex structure and geometry of the pores severely limit application of modern Finite Element Analysis methods because they require construction of spatial grids (meshes) that conform to the geometric shape of the structure. As a result, there are virtually no effective tools available for predicting overall mechanical and thermal properties of porous materials based on their microstructure. This thesis is a separate handling and controls of geometric and physical computational models that are seamlessly combined at solution run time. Using the proposedapproach we will determine the effective thermal conductivity tensor of real porous ceramic materials featuring both isotropic and anisotropic thermal properties. This work involved development and implementation of numerical algorithms, data structure, and software.