Physical Sciences and Mathematics Commons^™

High-Dimensional Indexing Methods Utilizing Clustering And Dimensionality Reduction, Lijuan Zhang

Dissertations

The emergence of novel database applications has resulted in the prevalence of a new paradigm for similarity search. These applications include multimedia databases, medical imaging databases, time series databases, DNA and protein sequence databases, and many others. Features of data objects are extracted and transformed into high-dimensional data points. Searching for objects becomes a search on points in the high-dimensional feature space. The dissimilarity between two objects is determined by the distance between two feature vectors. Similarity search is usually implemented as nearest neighbor search in feature vector spaces. The cost of processing k-nearest neighbor (k-NN) queries via a sequential …

Improving Document Representation By Accumulating Relevance Feedback : The Relevance Feedback Accumulation (Rfa) Algorithm, Razvan Stefan Bot

Dissertations

Document representation (indexing) techniques are dominated by variants of the term-frequency analysis approach, based on the assumption that the more occurrences a term has throughout a document the more important the term is in that document. Inherent drawbacks associated with this approach include: poor index quality, high document representation size and the word mismatch problem. To tackle these drawbacks, a document representation improvement method called the Relevance Feedback Accumulation (RFA) algorithm is presented. The algorithm provides a mechanism to continuously accumulate relevance assessments over time and across users. It also provides a document representation modification function, or document representation learning …

The Control Of Frequency Of A Conditional Oscillator Simultaneously Subjected To Multiple Oscillatory Inputs, Christina L. Ambrosio

Dissertations

A conditional oscillator is one that requires input to oscillate. An example of such is the gastric mill network of the stomatogastric ganglion of the crab Cancer borealis which requires modulatory input from outside the stomatogastric ganglion and fast input from the pyloric network of the animal in order to become active. This dissertation studies how the frequency of the gastric mill network is determined when it is simultaneously subjected to two different rhythmic inputs whose timing may be mismatched. We derive a mathematical model of the gastric mill network and deduce that the difference in timing between the pyloric …

Order Scheduling In Dedicated And Flexible Machine Environments, Haibing Li

Dissertations

Order scheduling models are relatively new in the field of scheduling. Consider a facility with m parallel machines that can process k different products (job types). Each machine can process a given subset of different product types. There are n orders from n different clients. Each order requests specific quantities of the various different products that can be produced concurrently on their given subsets of machines; it may have a release date, a weight and a due date. Preemptions may be allowed. An order can not be shipped until the processing of all the products for the order has been …

Flash Animation Project: Written Report, Peter Dee

Dissertations

A written report about a Flash animation project to include the 3 headings of Animation Techniques, Story Telling and Audio along with a critique and an appendix. The Flash animation project is based upon one of Aesop's fables entitled 'The King's Son & the Painted Lion'.

Structure Preserving Algorithms For Computing The Symplectic Singular Value Decom Position, Archara Chaiyakarn

Dissertations

In this thesis we develop two types of structure preserving Jacobi algorithms for com puting the symplectic singular value decomposition of real symplectic matrices and complex symplectic matrices. Unlike general purpose algorithms, these algorithms produce symplectic structure in all factors of the singular value decomposition.

Our first algorithm uses the relation between the singular value decomposition and the polar decomposition to reduce the problem of finding the symplectic singular value decomposition to th a t of calculating the structured spectral decomposition of a doubly structured m atrix. A Jacobi-like m ethod is developed to compute this doubly structured spectral decomposition. …

Stratification And Domination In Graphs And Digraphs, Ralucca M. Gera

Dissertations

In this thesis we combine the idea of stratification with the one of domination in graphs and digraphs, respectively.

A graph is 2-stratified if its vertex set is partitioned into two classes, where the vertices in one class are colored red and those in the other class are colored blue. Let F be a 2-stratified graph rooted at some blue vertex v . An F -coloring of a graph G is a red-blue coloring of the vertices of G in which every blue vertexu belongs to a copy of F rooted at u . The F -domination number γ …

Global Optimality Conditions In Mathematical Programming And Optimal Control, Pariwat Pacheenburawaa

Dissertations

We derive new first-order necessary and sufficient optimality conditions characterizing global minimizers in mathematical programming and optimal controlproblems. These conditions are based on level sets of an objective functional and they do not assume special structure of a problem (convexity, linearity, etc.). For a mathematical programming problem of minimization of a smooth functional on some compact convex set with equality nonlinear constraints, we derive first-order optimality conditions in the form of a generalized Lagrange multiplier rule. This rule should hold for any point from the level set of the objective functional corresponding to a global minimizer. We demonstrate that these …

Hybrid Explicit-Implicit Fdtd-Fem Time-Domain Solver For Electromagnetic Problems, Kakhkhor Abdijalilov

Dissertations

The Finite-Difference Time-Domain (FDTD) method and Finite-Element (FEM) method are numerical techniques used for solving Maxwell's electromagnetic equations. FDTD-FEM hybrid methods opt for combining the advantages of both FDTD and FEM. In this dissertation, signal processing techniques were used to analyze the FDTD stability condition. A procedure, which reduces time-sampling error yet preserves the stability of algorithm is proposed. Both explicit and implicit time-stepping schemes were treated in the framework of the developed method. An improved version of the implicit-explicit FEM-FDTD hybrid method was developed. The new method minimizes reflection from the interface between different types of grids. A class …

Characteristics Of Nanocomposites And Semiconductor Heterostructure Wafers Using Thz Spectroscopy, Hakan Altan

Dissertations

All optical, THz-Time Domain Spectroscopic (THz-TDS) methods were employed towards determining the electrical characteristics of Single Walled Carbon Nanotubes, Ion Implanted Si nanoclusters and Si_1-xGe_x HFO₂, SiO₂ on p-type Si wafers.

For the nanoscale composite materials, Visible Pump/THz Probe spectroscopy measurements were performed after observing that the samples were not sensitive to the THz radiation alone. The results suggest that the photoexcited nanotubes exhibit localized transport due to Lorentz-type photo-induced localized states from 0.2 to 0.7THz. The THz transmission is modeled through the photoexcited layer with an effective dielectric constant described by a Drude …

Negotiation Of Software Requirements In An Asynchronous Collaborative Environment, Catherine Lowry Campbell

Dissertations

The effect of task structure and negotiation sequence on collaborative software requirements negotiation is investigated. This work began with an extensive literature review that focused on current research in collaborative software engineering and, in particular, on the negotiation of software requirements and the requisite collaboration for the development of such requirements. A formal detailed experiment was then conducted to evaluate the effects of negotiation sequence and task structure in an asynchronous group meeting environment. The experiment tested the impact of these structures on groups negotiating the requirements for an emergency response information system. The results reported here show that these …

Characterizing The Evolution Of Operating Systems, Yi Peng

Dissertations

Examining the development and trends in software engineering technology is a huge undertaking. It is constantly evolving and affected by a large number of factors, which are themselves driven by a wide range of sub-factors. This dissertation is part of a long term project intended to analyze software engineering technology trends and how they evolve. This project is intended to analyze operating system trends and what are the factors that drive how they evolve. Basically, the following questions will be answered: "How to watch, predict, adapt to, and affect operating system's evolution trends?"

In previous research, YF Chen used statistical …

Live, Virtual, And Constructive Environments For Performance Support, John Michael Lacontora

Dissertations

As military systems become more complex, the operation and support of these systems becomes intrinsically more difficult. The U.S. Army's current procurement process relies on industry to provide embedded training and performance support tools for the systems they produce. These tools are relatively new and in the early stages of development. As yet, they have failed to meet the needs of the technicians that are required to support these complex systems. Current efforts to provide enabling technologies that enhance the capabilities of automotive maintenance technicians are concentrated in three professional communities. First is the Performance Improvement community where work is …

Pattern Discovery In Structural Databases With Applications To Bioinformatics, Sen Zhang

Dissertations

Frequent structure mining (FSM) aims to discover and extract patterns frequently occurring in structural data such as trees and graphs. FSM finds many applications in bioinformatics, XML processing, Web log analysis, and so on. In this thesis, two new FSM techniques are proposed for finding patterns in unordered labeled trees. Such trees can be used to model evolutionary histories of different species, among others.

The first FSM technique finds cousin pairs in the trees. A cousin pair is a pair of nodes sharing the same parent, the same grandparent, or the same great-grandparent, etc. Given a tree T, our …

The Testing Of Semiconductor-Based Adsorption Modified Photosensitive Sensors For Their Response To A Volatile Organic Compound, Oxygen, Humidity, And Heating, Roberta Rosty

Dissertations

Two thin-film sensors, composed of different thicknesses (approximately 0.5 microns and 4.5 microns) of a cadmium sulfide layer coated with Rhodamine B fluorescent dye on a glass substrate, were tested for change in photoconductivity due to exposure to different concentrations of gaseous organic molecules in the parts-per-million (ppm) range.

It was theorized that the gaseous organic molecules would adsorb to the dyed semiconductor surface and that some energy would be transferred to the adsorbed analyte rather than to the cadmium sulfide semiconductor layer through the dye, thereby decreasing the photoconductivity of the surface in an amount proportional to the analyte …

Scattering Matrix Analysis Of Photonic Crystals, Valeriy Lukyanov

Dissertations

Using a scattering matrix approach we analyze and study the scattering and transmission of waves through a two-dimensional photonic crystal which consists of a periodic array of parallel rods with circular cross sections. Without making any assumptions about normal incidence, single mode propagation, and sufficient inter-scatter separation in the direction of propagation, we show how to compute the transmission and reflection coefficients of these periodic structures. The method is based on the computation of a generalized scattering matrix for one column of the periodic structure.

We also develop an analytical method to analyze and to study the scattering and transmission …

An Optic Fiber Sensor For Partial Discharge Acoustic Detection, Xiaodong Wang

Dissertations

Partial discharge (PD) is a very common problem in operating power transformers and is one of the factors that could lead to failure of power transformers, leading to power outage and expensive repairs. The acoustic wave induced by PD can be measured and used for monitoring, diagnosing, and locating potential failures in power transformers.

The effects of the temperature of the transformer and transformer oil are one of the very important parameters in PD and these effects are investigated in detail. The Fast Fourier Transform (FF1') is used to synthesize the measured data and results show that for periodic PD …