Physical Sciences and Mathematics Commons^™

Performance Comparison Of Two Data Mining Algorithms On Big Data Platforms, Md Rajiur Rahman Raju

Wayne State University Theses

In this Big data era, the need for performing large-scale computations is evident. A better understanding of the most suitable platforms which can efficiently run these computations is needed. In this thesis, we attempt to compare four such big data platforms, namely Hadoop, Spark, GPU, and Multicore CPU. We compare these platforms using two prominent data mining algorithms, namely, K-means clustering and K-nearest neighbour classification and discuss specific implementation-level details. We provide several insights into the best possible implementations of these algorithms and systematically compare the benefits and drawbacks of each of these platforms. We conduct experiments by varying data …

The Impact Of Increased Optimization Problem Dimensionality On Cultural Algorithm Performance, Yang Yang

Wayne State University Theses

ABSTRACT

The Impact of Increased Optimization Problem Dimensionality on

Cultural Algorithm Performance

by

Yang Yang

August 2015

Advisor: Dr. Robert Reynolds

Major: Computer Science

Degree: Master of Science

In this thesis, we investigate the performance of Cultural Algorithms when dealing with the increasing dimensionality of optimization problems. The research is based on previous cultural algorithm approaches with the Cultural Algorithms Toolkit, CAT 2.0, which supports a variety of co-evolutionary features at both the knowledge and population levels. In this project, the system was applied to the solution of 60 randomly generated problems that ranged from 2-dimensional to 5-dimensional problem spaces. …

Shape Analysis Using Spectral Geometry, Jiaxi Hu

Wayne State University Dissertations

Shape analysis is a fundamental research topic in computer graphics and computer vision. To date, more and more 3D data is produced by those advanced acquisition capture devices, e.g., laser scanners, depth cameras, and CT/MRI scanners. The increasing data demands advanced analysis tools including shape matching, retrieval, deformation, etc. Nevertheless, 3D Shapes are represented with Euclidean transformations such as translation, scaling, and rotation and digital mesh representations are irregularly sampled. The shape can also deform non-linearly and the sampling may vary. In order to address these challenging problems, we investigate Laplace-Beltrami shape spectra from the differential geometry perspective, focusing more …

Predictable Real-Time Wireless Networking For Sensing And Control, Xiaohui Liu

Wayne State University Dissertations

Towards the end goal of providing predictable real-time wireless networking for sensing and control, we have developed a real-time routing protocol MTA that predictably delivers data by their deadlines, and a scheduling protocol PRKS to ensure a certain link reliability based on the Physical-ratio-K (PRK) model, which is both realistic and amenable for distributed implementation, and a greedy scheduling algorithm to deliver as many packets as possible to the sink by a deadline in lossy multi-hop wireless sensor networks.

Real-time routing is a basic element of closed-loop, real-time sensing and control, but it is challenging due to dynamic, uncertain link/path …

Evolving Heterogeneous And Subcultured Social Networks For Optimization Problem Solving In Cultural Algorithms, Yousof Gawasmeh

Wayne State University Dissertations

Cultural Algorithms are computational models of social evolution based upon principle of Cultural Evolution. A Cultural Algorithm are composed of a Belief Space consisting of a network of active and passive knowledge sources and a Population Space of agents. The agents are connected via a social fabric over which information used in agent problem solving is passed. The knowledge sources in the Belief Space compete with each other in order to influence the decision making of agents in the Population Space. Likewise, the problem solving experiences of agents in the Population Space are sent back to the Belief Space and …

Semantic Web Based Relational Database Access With Conflict Resolution, Fayez Khazalah

Wayne State University Dissertations

This thesis focuses on (1) accessing relational databases through Semantic Web technologies and (2) resolving conflicts that usually arises when integrating data from heterogeneous source schemas and/or instances.

In the first part of the thesis, we present an approach to access relational databases using Semantic Web technologies. Our approach is built on top of Ontop framework for Ontology Based Data Access. It extracts both Ontop mappings and an equivalent OWL ontology from an existing database schema. The end users can then access the underlying data source through SPARQL queries. The proposed approach takes into consideration the different relationships between the …

Applications Of Machine Learning In Biology And Medicine, Saied Haidarian Shahri

Wayne State University Dissertations

Machine learning as a field is defined to be the set of computational algorithms that improve their performance by assimilating data.

As such, the field as a whole has found applications in many diverse disciplines from robotics and communication in engineering to economics and finance, and also biology and medicine.

It should not come as a surprise that many popular methods in use today have completely different origins.

Despite this heterogeneity, different methods can be divided into standard tasks, such as supervised, unsupervised, semi-supervised and reinforcement learning.

Although machine learning as a field can be formalized as methods trying to …

Resource Management In Cloud And Big Data Systems, Lena Mashayekhy

Wayne State University Dissertations

Cloud computing is a paradigm shift in computing, where services are offered and acquired on demand in a cost-effective way. These services are often virtualized, and they can handle the computing needs of big data analytics. The ever-growing demand for cloud services arises in many areas including healthcare, transportation, energy systems, and manufacturing. However, cloud resources such as computing power, storage, energy, dollars for infrastructure, and dollars for operations, are limited. Effective use of the existing resources raises several fundamental challenges that place the cloud resource management at the heart of the cloud providers' decision-making process. One of these challenges …

Efficient Synergistic De Novo Co-Assembly Of Bacterial Genomes From Single Cells Using Colored De Bruijn Graph, Narjes Sadat Movahedi Tabrizi

Wayne State University Dissertations

Recent progress in DNA amplification techniques, particularly multiple displacement

amplification (MDA), has made it possible to sequence and assemble bacterial

genomes from a single cell. However, the quality of single cell genome assembly has

not yet reached the quality of normal multi-cell genome assembly due to the coverage

bias (including uneven depth of coverage and region blackout) and errors caused by

MDA. Computational methods try to mitigates the amplification bias. In this document

we introduce a de novo co-assembly method using colored de Bruijn graph,

which can overcome the problem of blackout regions due to amplification bias. The

algorithm is …

Building Computing-As-A-Service Mobile Cloud System, Kun Wang

Wayne State University Dissertations

The last five years have witnessed the proliferation of smart mobile devices, the explosion of various mobile applications and the rapid adoption of cloud computing in business, governmental and educational IT deployment. There is also a growing trends of combining mobile computing and cloud computing as a new popular computing paradigm nowadays. This thesis envisions the future of mobile computing which is primarily affected by following three trends: First, servers in cloud equipped with high speed multi-core technology have been the main stream today. Meanwhile, ARM processor powered servers is growingly became popular recently and the virtualization on ARM systems …

Efficient Algorithms And Optimizations For Scientific Computing On Many-Core Processors, Kamel Rushaidat

Wayne State University Dissertations

Designing efficient algorithms for many-core and multicore architectures requires using different strategies to allow for the best exploitation of the hardware resources on those architectures. Researchers have ported many scientific applications to modern many-core and multicore parallel architectures, and by doing so they have achieved significant speedups over running on single CPU cores. While many applications have achieved significant speedups, some applications still require more effort to accelerate due to their inherently serial behavior.

One class of applications that has this serial behavior is the Monte Carlo simulations. Monte Carlo simulations have been used to simulate many problems in statistical …

Object Tracking: Appearance Modeling And Feature Learning, Raed Almomani

Wayne State University Dissertations

Object tracking in real scenes is an important problem in computer vision due to increasing usage of tracking systems day in and day out in various applications such as surveillance, security, monitoring and robotic vision. Object tracking is the process of locating objects of interest in every frame of video frames. Many systems have been proposed to address the tracking problem where the major challenges come from handling appearance variation during tracking caused by changing scale, pose, rotation, illumination and occlusion.

In this dissertation, we address these challenges by introducing several novel tracking techniques. First, we developed a multiple object …

A Control-Theoretic Design And Analysis Framework For Resilient Hard Real-Time Systems, Pradeep Mahendra Hettiarachchi

Wayne State University Dissertations

We introduce a new design metric called system-resiliency which characterizes the maximum unpredictable

external stresses that any hard-real-time performance mode can withstand. Our proposed systemresiliency

framework addresses resiliency determination for real-time systems with physical and hardware

limitations. Furthermore, our framework advises the system designer about the feasible trade-offs between

external system resources for the system operating modes on a real-time system that operates in a

multi-parametric resiliency environment.

Modern multi-modal real-time systems degrade the system’s operational modes as a response to unpredictable

external stimuli. During these mode transitions, real-time systems should demonstrate a reliable

and graceful degradation of service. Many …

Well-Posedness Properties In Variational Analysis And Its Applications, Wei Ouyang

Wayne State University Dissertations

This dissertation focuses on the study and applications of some significant properties in well-posedness and sensitivity analysis, among which the notions of uniform metric regularity , higher-order metric subregularity and its strong subregularity counterpart play an essential role in modern variational analysis. We derived verifiable sufficient conditions and necessary conditions for those notions in terms of appropriate generalized differential as well as geometric constructions of variational analysis. Concrete examples are provided to illustrate the behavior and compare the results. Optimality conditions of parametric variational systems (PVS) under equilibrium constraints are also investigated via the terms of coderivatives. We derived necessary …

Molecular Physics Of Elementary Processes Relevant To Hypersonics: Electron-Molecule Collisions, R Celiberto, V Laporta, A Laricchiuta, J Tennyson, J. M. Wadehra

Physics and Astronomy Faculty Research Publications

Non-resonant, electron-impact, vibro-electronic excitation cross sections, involving vibrationally excited N2 molecules, to the mixed valence-Rydberg b,c,o 1Πu and b′,c′,e′ 1åu+ singlet states are presented. These cross sections are calculated using the so-called similarity approach, accounting for the vibronic coupling among excited states, and compared with the experiments and different theoretical calculations.

New cross sections for the electron-impact resonant vibrational excitation of CO2 molecule are calculated, for the symmetric stretching mode, as a function of the incident electron energy and for the transitions (υ i , 0,0)→(νf , 0,0) with νi = 0,1,2 and for some selected value of νf in …

Classical And Motivic Adams Charts, Daniel C. Isaksen

Mathematics Research Reports

This document contains large-format Adams charts that compute 2-complete stable homotopy groups, both in the classical context and in the motivic context over C. The charts are essentially complete through the 59-stem and contain partial results to the 70-stem. In the classical context, we believe that these are the most accurate charts of their kind. We also include Adams charts for the motivic homotopy groups of the cofiber of τ.

Classical And Motivic Adams-Novikov Charts, Daniel C. Isaksen

Mathematics Research Reports

This document contains large-format Adams-Novikov charts that compute the classical 2-complete stable homotopy groups. The charts are essentially complete through the 59-stem. We believe that these are the most accurate and extensive charts of their kind. We also include a motivic Adams-Novikov E∞ chart.

Retained-Components Factor Transformation: Factor Loadings And Factor Score Predictors In The Column Space Of Retained Components, André Beauducel, Frank Spohn

Journal of Modern Applied Statistical Methods

Factor loadings optimally account for the non-diagonal elements of the covariance matrix of observed variables. Principal component analysis leads to components accounting for a maximum of the variance of the observed variables. Retained-components factor transformation is proposed in order to combine the advantages of factor analysis and principal component analysis.

Pairwise Comparison In Repeated Measures, I.C.A. Oyeka, C. C. Nnanatu

Journal of Modern Applied Statistical Methods

Sometimes a random sample of subjects or patients may be exposed to a battery of diagnostic tests or medication over time and interest is on determining whether there is progressive remission of condition, disease or symptom. Also perhaps early in a program or experiment, subjects or candidates may be required to significantly improve in their performance rates at the current trial relative to an immediately preceding trial, otherwise they may have to withdraw from or drop out. The research interest would then be to determine some critical minimum marginal success rate to guide the management in decision making as well …

A Comparison Of Methods For Group Prediction With High Dimensional Data, Holmes Finch

Journal of Modern Applied Statistical Methods

High dimensional data is the situation in which the number of variables included in an analysis approaches or exceeds the sample size. In the context of group classification, researchers are typically interested in finding a model that can be used to correctly place an individual into their appropriate group; e.g. correctly diagnose individuals with depression. However, when the size of the training sample is small and the number of predictors used to differentiate the groups is larger, standard approaches such as discriminant analysis may not work well. In order to address this issue, statisticians have developed a number of tools …

Objective Priors For Estimation Of Extended Exponential Geometric Distribution, Pedro L. Ramos, Fernando A. Moala, Jorge A. Achcar

Journal of Modern Applied Statistical Methods

A Bayesian analysis was developed with different noninformative prior distributions such as Jeffreys, Maximal Data Information, and Reference. The aim was to investigate the effects of each prior distribution on the posterior estimates of the parameters of the extended exponential geometric distribution, based on simulated data and a real application.

Some General Guidelines For Choosing Missing Data Handling Methods In Educational Research, Jehanzeb R. Cheema

Journal of Modern Applied Statistical Methods

The effect of a number of factors, such as the choice of analytical method, the handling method for missing data, sample size, and proportion of missing data, were examined to evaluate the effect of missing data treatment on accuracy of estimation. A methodological approach involving simulated data was adopted. One outcome of the statistical analyses undertaken in this study is the formulation of easy-to-implement guidelines for educational researchers that allows one to choose one of the following factors when all others are given: sample size, proportion of missing data in the sample, method of analysis, and missing data handling method.

Bayesian Inference For Volatility Of Stock Prices, Juliet G. D'Cunha, K. A. Rao

Journal of Modern Applied Statistical Methods

Lognormal distribution is widely used in the analysis of failure time data and stock prices. Maximum likelihood and Bayes estimator of the coefficient of variation of lognormal distribution along with confidence/credible intervals are developed. The utility of Bayes procedure is illustrated by analyzing prices of selected stocks.

Local Bandwidths For Improving Performance Statistics Of Model-Robust Regression 2, Efosa Edionwe, Julian L. Mbegbu

Journal of Modern Applied Statistical Methods

Model-Robust Regression 2 (MRR2) method is a semi-parametric regression approach that combines parametric and nonparametric fits. The bandwidth controls the smoothness of the nonparametric portion. We present a methodology for deriving data-driven local bandwidth that enhances the performance of MRR2 method for fitting curves to data generated from designed experiments.

Contrast Of Bayesian And Classical Sample Size Determination, Farhana Sadia, Syed S. Hossain

Journal of Modern Applied Statistical Methods

Sample size determination is a prerequisite for statistical surveys. A comprehensive overview of the Bayesian approach for computation of the sample size, and a comparison with classical approaches, is presented. Two surveys are taken as example to illustrate the accuracy and efficiency of each approach, and to make recommendations about which method is preferred. The Bayesian approach of sample size determination may require fewer subjects if proper prior information is available.

The Information Criterion, Masume Ghahramani

Journal of Modern Applied Statistical Methods

The Akaike information criterion, AIC, is widely used for model selection. Using the AIC as the estimator of asymptotic unbias for the second term Kullbake-Leibler risk considers the divergence between the true model and offered models. However, it is an inconsistent estimator. A proposed approach the problem is the use of A'IC, a consistently offered information criterion. Model selection of classic and linear models are considered by a Monte Carlo simulation.

Gumbel-Weibull Distribution: Properties And Applications, Raid Al-Aqtash, Carl Lee, Felix Famoye

Journal of Modern Applied Statistical Methods

Some properties of the Gumbel-Weibull distribution including the mean deviations and modes are studied. A detailed discussion of regions of unimodality and bimodality is given. The method of maximum likelihood is proposed for estimating the distribution parameters and a simulation is conducted to study the performance of the method. Three tests are given for testing the significance of a distribution parameter. The applications of Gumbel-Weibull distribution are emphasized. Five data sets are used to illustrate the flexibility of the distribution in fitting unimodal and bimodal data sets.

Fitting Stereotype Logistic Regression Models For Ordinal Response Variables In Educational Research (Stata), Xing Liu

Journal of Modern Applied Statistical Methods

The stereotype logistic (SL) model is an alternative to the proportional odds (PO) model for ordinal response variables when the proportional odds assumption is violated. This model seems to be underutilized. One major reason is the constraint of current statistical software packages. Statistical Package for the Social Sciences (SPSS) cannot perform the SL regression analysis, and SAS does not have the procedure developed to directly estimate the model. The purpose of this article was to illustrate the stereotype logistic (SL) regression model, and apply it to estimate mathematics proficiency level of high school students using Stata. In addition, it compared …

Optimal Location Design For Prediction Of Spatial Correlated Environmental Functional Data, Mahdi Rasekhi, B. Jamshidi, F. Rivaz

Journal of Modern Applied Statistical Methods

The optimal choice of sites to make spatial prediction is critical for a better understanding of really spatio-temporal data. It is important to obtain the essential spatio-temporal variability of the process in determining optimal design, because these data tend to exhibit both spatial and temporal variability. Two new methods of prediction for spatially correlated functional data are considered. The first method models spatial dependency by fitting variogram to empirical variogram, similar to ordinary kriging (univariate approach). The second method models spatial dependency by linear model co-regionalization (multivariate approach). The variance of prediction method was chosen as the optimization design criterion. …

Missing Data And The Statistical Modeling Of Adolescent Pregnancy, Dudley L. Poston Dr., Eugenia Conde Dr.

Journal of Modern Applied Statistical Methods

Missing data is a pervasive problem in social science research. Many techniques have been developed to handle the problem. Different ways of handling missing data were shown to lead to different results in statistical models. A demonstration was given based on statistical modeling of the likelihood of a woman reporting having had an adolescent pregnancy by handling missing data with several different approaches. Results indicate that many of the independent variables in the model vary in whether they are, or are not, statistically significant in predicting the log odds of a woman having a teen pregnancy, and in the ranking …