The present thesis is based on three studies that research different aspects of fine texture perception. The goal is to better understand the mechanisms involved in haptic perception of textures below 200 µm, also known as microtextures. Study I was conducted to establish a friction measurement model and relating the friction measurements to perceived coarseness of fine textures. A set of printing papers was used as stimulus material. In Study II an expanded set, including the set of Study I, was used as stimuli in a multidimensional scaling (MDS) experiment of haptic fine texture perception. Through scaling of perceptual attributes and similarities, a three dimensional space was found to best describe the data and the dimensions were interpreted as rough-smooth, thick-thin and distinct-indistinct. In Study III a series of model surfaces were manufactured with a systematically varied sinusoidal pattern, spanning from 300 nm to 80 µm. As in Study II, a similarity experiment was conducted and a two dimensional space was chosen, the dimensions of which were explained well through friction and the wavelength. Together these three studies form a better picture of fine texture perception. The dimensionality found with paper stimuli was very similar to the corresponding spaces for marcrotextures of everyday materials, even though a different perceptual system is used for fine texture perception. Regardless if the information is coded through the spatial or the vibratory sense, the perception does not seem to differ in dimensionality. Further, the largest among the microtextures seem to have been perceived as carrying spatial information. On the systematically varied, rigid, textures, the MDS space did not come out in a similar fashion to those of everyday materials but instead similar to the physical properties that characterizes the change in the textures. It was further found that the participants in Study III successfully discriminated textures with an amplitude of 13 nm from the unwrinkled surfaces. From these studies the main conclusions are (a) haptically measured friction and surface roughness are important contributors to fine texture perception, (b) even at microscales, spatial information is retrieved haptically, probably through vibrations, and (c) persons can haptically discriminate textures at a nanoscale.

Social networks have been evolving over the past few years, leading to a rapid increase in the number and complexity of relationships among their entities. In this research, we focus on a large scale dataset known as the Digital Bibliography and Library Project or DBLP, which contains information on all publications that have been published in computer and information science related journals and conference proceedings. We model the DBLP dataset as a social network of research collaborations. DBLP is a structured and dynamic dataset stored in the XML file format; it contains over 850,000 authors and 2 million publications, and the resulting collaboration social network is a scale-free network. We define DBLP collaboration social network as a graph that consists of researchers as nodes and links representing the collaboration or co-authorship relationships among the researchers. In this work, we implement a data analysis algorithm called Multidimensional Scaling (MDS) to represent the degree of collaboration among the DBLP authors as Euclidean distances in 2-dimensional space in order to analyze, mine and understand the relational information in this large scale network in a visual way. MDS is a useful technique for data visualization and graph drawing methods, but it has high computational complexity for large scale graphs such as the DBLP graph. Therefore, we propose different solutions to overcome this problem, and improve the MDS performance. In addition, as the quality of the MDS result is measured by a metric known as the stress value, we use the steepest descent method to minimize the stress in an iterative process called stress optimization in order to generate the best geometric layout of the graph nodes in 2-dimensional space. We also propose a solution to further enhance the graph visualization by partitioning the graph into sub-graphs and using repelling forces among nodes within the same sub-graph. Finally, we developed a new visualization tool that can handle the large scale of the DBLP graph, and provides the user a number of significant features that allow them to explore, navigate and sift for information through the graph, such as graph scaling and graphical search functionality. / UOIT

The ability to visualize a solution space can be very beneficial, and it is generally accepted that the objective of visualization is to aid researchers in gathering insight. However, insight cannot be gathered effectively if the source data is misrepresented. This dissertation begins by demonstrating that the adaptive landscape visualization in widespread usage frequently misrepresents the neighborhood structure of genotypic space and, consequently, will mislead users about the manner in which solution space is traversed by the genetic algorithm. Bernhard Riemann, the father of topology, explicitly noted that a measurement of the distance between entities should represent the manner in which one can be brought towards the other. Thus, the commonly used Hamming distance, for example, is not representative of traversals of genotypic space by the genetic algorithm – a representative measure must include consideration for both mutation and recombination. This dissertation separately explores the properties that mutational and recombinational distances should have, and ultimately establishes a measure that is representative of the traversals made by both operators simultaneously. It follows that these measures can be used to enhance the adaptive landscape, by minimizing the discrepancy between the interpoint distances in genotypic space and the interpoint distances in the two-dimensional representation from which the landscape is extruded. This research also establishes a methodology for evaluating measures defining neighbourhood structures that are purportedly representative of traversals of genotypic space, by comparing them against an empirically generated norm. Through this approach it is conclusively demonstrated that the Hamming distance between genotypes is less representative than the proposed measures, and should not be used to define the neighbourhood structure from which visualizations would be constructed. While the proposed measures do not distort the data or otherwise mislead the user, they do require a significant computational expense. Fortunately, the choice to use these measures is always made at the discretion of the user, with additional costs incurred when accuracy and representativity are of paramount importance. These measures will ultimately find further application in population diversity measurement, cluster analysis, and any other task where the representativity of the neighborhood structure of the genotypic space is vital.

No description available.

No description available.

Thesis (M.S.)--University of Missouri-Columbia, 2007. / The entire dissertation/thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file (which also appears in the research.pdf); a non-technical general description, or public abstract, appears in the public.pdf file. Title from title screen of research.pdf file (viewed on April 17, 2008) Includes bibliographical references.

Thesis (Ph.D.)--Michigan State University. Measurement and Quantiative Methods Educational Policy, 2008. / Title from PDF t.p. (viewed on Mar. 30, 2009) Includes bibliographical references (p.193-199). Also issued in print.

Thesis (MComm)--University of Stellenbosch, 2008. / Bibliography. Also available via the Internet.

Principal surfaces are smooth two-dimensional surfaces that pass through the middle of a p-dimensional data set. They minimise the distance from the data points, and provide a nonlinear summary of the data. The surfaces are nonparametric and their shape is suggested by the data. The formation of a surface is found using an iterative procedure which starts with a linear summary, typically with a principal component plane. Each successive iteration is a local average of the p-dimensional points, where an average is based on a projection of a point onto the nonlinear surface of the previous iteration. Biplots are considered as extensions of the ordinary scatterplot by providing for more than three variables. When the difference between data points are measured using a Euclidean embeddable dissimilarity function, observations and the associated variables can be displayed on a nonlinear biplot. A nonlinear biplot is predictive if information on variables is added in such a way that it allows the values of the variables to be estimated for points in the biplot. Prediction trajectories, which tend to be nonlinear are created on the biplot to allow information about variables to be estimated. The goal is to extend the idea of nonlinear biplot methodology onto principal surfaces. The ultimate emphasis is on high dimensional data where the nonlinear biplot based on a principal surface allows for visualisation of samples, variable trajectories and predictive sets of contour lines. The proposed biplot provides more accurate predictions, with an additional feature of visualising the extent of nonlinearity that exists in the data.

The purpose of this quantitative study was to further explore pedagogical balance using multidimensional scaling and epistemic network analysis. Teacher shortages and attrition remain a critical issue for the future, and simulated classroom environments like simSchool can provide teachers additional training to help improve teaching confidence and teaching experience. Two different data sets were analyzed at various time before and after simSchool use to see how pre-service teachers rate themselves in 8 areas of teaching as defined by the Survey of Teaching Skills. Multidimensional scaling was utilized to see how teaching confidence and teaching experience align with no simSchool use, 90 minutes of use, and 8 hours of use. Epistemic network analysis was utilized to look at the cognitive structures of different groups to determine any differences. The findings are discussed with future research directions provided.