Computer Sciences Commons^™

Spatiotemporal Subspace Feature Tracking By Mining Discriminatory Characteristics, Richard D. Appiah

Doctoral Dissertations

Recent advancements in data collection technologies have made it possible to collect heterogeneous data at complex levels of abstraction, and at an alarming pace and volume. Data mining, and most recently data science seek to discover hidden patterns and insights from these data by employing a variety of knowledge discovery techniques. At the core of these techniques is the selection and use of features, variables or properties upon which the data were acquired to facilitate effective data modeling. Selecting relevant features in data modeling is critical to ensure an overall model accuracy and optimal predictive performance of future effects. The …

Motion-Capture-Based Hand Gesture Recognition For Computing And Control, Andrew Gardner

Doctoral Dissertations

This dissertation focuses on the study and development of algorithms that enable the analysis and recognition of hand gestures in a motion capture environment. Central to this work is the study of unlabeled point sets in a more abstract sense. Evaluations of proposed methods focus on examining their generalization to users not encountered during system training.

In an initial exploratory study, we compare various classification algorithms based upon multiple interpretations and feature transformations of point sets, including those based upon aggregate features (e.g. mean) and a pseudo-rasterization of the capture space. We find aggregate feature classifiers to be balanced across …

A Framework For The Statistical Analysis Of Mass Spectrometry Imaging Experiments, Kyle Bemis

Open Access Dissertations

Mass spectrometry (MS) imaging is a powerful investigation technique for a wide range of biological applications such as molecular histology of tissue, whole body sections, and bacterial films , and biomedical applications such as cancer diagnosis. MS imaging visualizes the spatial distribution of molecular ions in a sample by repeatedly collecting mass spectra across its surface, resulting in complex, high-dimensional imaging datasets. Two of the primary goals of statistical analysis of MS imaging experiments are classification (for supervised experiments), i.e. assigning pixels to pre-defined classes based on their spectral profiles, and segmentation (for unsupervised experiments), i.e. assigning pixels to newly …

Computational Environment For Modeling And Analysing Network Traffic Behaviour Using The Divide And Recombine Framework, Ashrith Barthur

Open Access Dissertations

There are two essential goals of this research. The first goal is to design and construct a computational environment that is used for studying large and complex datasets in the cybersecurity domain. The second goal is to analyse the Spamhaus blacklist query dataset which includes uncovering the properties of blacklisted hosts and understanding the nature of blacklisted hosts over time.

The analytical environment enables deep analysis of very large and complex datasets by exploiting the divide and recombine framework. The capability to analyse data in depth enables one to go beyond just summary statistics in research. This deep analysis is …

Divide And Recombined For Large Complex Data: Nonparametric-Regression Modelling Of Spatial And Seasonal-Temporal Time Series, Xiaosu Tong

Open Access Dissertations

In the first chapter of this dissertation, I briefly introduce one type of nonparametric regression method, namely local polynomial regression, followed by emphasis on one specific application of loess on time series decomposition, called Seasonal Trend Loess (STL). The chapter is closed by the introduction of D\&R; (Divide and Recombined) statistical framework. Data can be divided into subsets, each of which is applied with a statistical analysis method. This is an embarrassing parallel procedure since there is no communication between each subset. Then the analysis result for each subset are combined together to be the final analysis outcome for the …

Controlling For Confounding Network Properties In Hypothesis Testing And Anomaly Detection, Timothy La Fond

Open Access Dissertations

An important task in network analysis is the detection of anomalous events in a network time series. These events could merely be times of interest in the network timeline or they could be examples of malicious activity or network malfunction. Hypothesis testing using network statistics to summarize the behavior of the network provides a robust framework for the anomaly detection decision process. Unfortunately, choosing network statistics that are dependent on confounding factors like the total number of nodes or edges can lead to incorrect conclusions (e.g., false positives and false negatives). In this dissertation we describe the challenges that face …

User-Centric Workload Analytics: Towards Better Cluster Management, Suhas Raveesh Javagal

Open Access Theses

Effective management of computing clusters and providing a high quality customer support is not a trivial task. Due to rise of community clusters there is an increase in the diversity of workloads and the user demographic. Owing to this and privacy concerns of the user, it is difficult to identify performance issues, reduce resource wastage and understand implicit user demands. In this thesis, we perform in-depth analysis of user behavior, performance issues, resource usage patterns and failures in the workloads collected from a university-wide community cluster and two clusters maintained by a government lab. We also introduce a set of …

Implementation And Validation Of A Probabilistic Open Source Baseball Engine (Posbe): Modeling Hitters And Pitchers, Rhett Tracy Schaefer

Open Access Theses

This manuscript details the implementation and validation of an open source probabilistic baseball engine (POSBE) that focuses on the hitter and pitcher model of the simulation. The simulation produced outcomes that parallel those observed in actual professional Major League Baseball games. The observed data were taken from the nineteen games played between the New York Yankees (NYY) and Boston Red Sox (BOS) during the 2015 season. The potential hitter/pitcher outcomes of interest were singles, doubles, triples, homeruns, walks, hit-by-pitch, and strikeouts. The nineteen game series was simulated 1000 times, resulting in a total of 19,000 simulations. The eighteen hitters and …

Sensitivity Of Mixed Models To Computational Algorithms Of Time Series Data, Gunaime Nevine

Doctoral Dissertations

Statistical analysis is influenced by implementation of the algorithms used to execute the computations associated with various statistical techniques. Over many years; very important criteria for model comparison has been studied and examined, and two algorithms on a single dataset have been performed numerous times. The goal of this research is not comparing two or more models on one dataset, but comparing models with numerical algorithms that have been used to solve them on the same dataset.

In this research, different models have been broadly applied in modeling and their contrasting which are affected by the numerical algorithms in different …

Performance Modeling And Optimization Techniques For Heterogeneous Computing, Supada Laosooksathit

Doctoral Dissertations

Since Graphics Processing Units (CPUs) have increasingly gained popularity amoung non-graphic and computational applications, known as General-Purpose computation on GPU (GPGPU), CPUs have been deployed in many clusters, including the world's fastest supercomputer. However, to make the most efficiency from a GPU system, one should consider both performance and reliability of the system.

This dissertation makes four major contributions. First, the two-level checkpoint/restart protocol that aims to reduce the checkpoint and recovery costs with a latency hiding strategy in a system between a CPU (Central Processing Unit) and a GPU is proposed. The experimental results and analysis reveals some benefits, …

Vulnerability Analysis Of Cyber-Behavioral Biometric Authentication, Abdul Serwadda

Doctoral Dissertations

Research on cyber-behavioral biometric authentication has traditionally assumed naïve (or zero-effort) impostors who make no attempt to generate sophisticated forgeries of biometric samples. Given the plethora of adversarial technologies on the Internet, it is questionable as to whether the zero-effort threat model provides a realistic estimate of how these authentication systems would perform in the wake of adversity. To better evaluate the efficiency of these authentication systems, there is need for research on algorithmic attacks which simulate the state-of-the-art threats.

To tackle this problem, we took the case of keystroke and touch-based authentication and developed a new family of algorithmic …

Automating Construction And Selection Of A Neural Network Using Stochastic Optimization, Jason Lee Hurt

UNLV Theses, Dissertations, Professional Papers, and Capstones

An artificial neural network can be used to solve various statistical problems by approximating a function that provides a mapping from input to output data. No universal method exists for architecting an optimal neural network. Training one with a low error rate is often a manual process requiring the programmer to have specialized knowledge of the domain for the problem at hand.

A distributed architecture is proposed and implemented for generating a neural network capable of solving a particular problem without specialized knowledge of the problem domain. The only knowledge the application needs is a training set that the network …

Machine Learning Approaches For Determining Effective Seeds For K -Means Algorithm, Kaveephong Lertwachara

Doctoral Dissertations

In this study, I investigate and conduct an experiment on two-stage clustering procedures, hybrid models in simulated environments where conditions such as collinearity problems and cluster structures are controlled, and in real-life problems where conditions are not controlled. The first hybrid model (NK) is an integration between a neural network (NN) and the k-means algorithm (KM) where NN screens seeds and passes them to KM. The second hybrid (GK) uses a genetic algorithm (GA) instead of the neural network. Both NN and GA used in this study are in their simplest-possible forms.

In the simulated data sets, I investigate two …

Bottom-Up Design Of Artificial Neural Network For Single-Lead Electrocardiogram Beat And Rhythm Classification, Srikanth Thiagarajan

Doctoral Dissertations

Performance improvement in computerized Electrocardiogram (ECG) classification is vital to improve reliability in this life-saving technology. The non-linearly overlapping nature of the ECG classification task prevents the statistical and the syntactic procedures from reaching the maximum performance. A new approach, a neural network-based classification scheme, has been implemented in clinical ECG problems with much success. The focus, however, has been on narrow clinical problem domains and the implementations lacked engineering precision. An optimal utilization of frequency information was missing. This dissertation attempts to improve the accuracy of neural network-based single-lead (lead-II) ECG beat and rhythm classification. A bottom-up approach defined …