Search Results (1 - 20 of 20 Results)

Sort By  
Sort Dir
 
Results per page  

Abounia Omran, BehzadApplication of Data Mining and Big Data Analytics in the Construction Industry
Doctor of Philosophy, The Ohio State University, 2016, Food, Agricultural and Biological Engineering
In recent years, the digital world has experienced an explosion in the magnitude of data being captured and recorded in various industry fields. Accordingly, big data management has emerged to analyze and extract value out of the collected data. The traditional construction industry is also experiencing an increase in data generation and storage. However, its potential and ability for adopting big data techniques have not been adequately studied. This research investigates the trends of utilizing big data techniques in the construction research community, which eventually will impact construction practice. For this purpose, the application of 26 popular big data analysis techniques in six different construction research areas (represented by 30 prestigious construction journals) was reviewed. Trends, applications, and their associations in each of the six research areas were analyzed. Then, a more in-depth analysis was performed for two of the research areas including construction project management and computation and analytics in construction to map the associations and trends between different construction research subjects and selected analytical techniques. In the next step, the results from trend and subject analysis were used to identify a promising technique, Artificial Neural Network (ANN), for studying two construction-related subjects, including prediction of concrete properties and prediction of soil erosion quantity in highway slopes. This research also compared the performance and applicability of ANN against eight predictive modeling techniques commonly used by other industries in predicting the compressive strength of environmentally friendly concrete. The results of this research provide a comprehensive analysis of the current status of applying big data analytics techniques in construction research, including trends, frequencies, and usage distribution in six different construction-related research areas, and demonstrate the applicability and performance level of selected data analytics techniques with an emphasis on ANN in construction-related studies. The main purpose of this dissertation was to help practitioners and researchers identify a suitable and applicable data analytics technique for their specific construction/research issue(s) or to provide insights into potential research directions.

Committee:

Qian Chen, Dr. (Advisor)

Subjects:

Civil Engineering; Comparative Literature; Computer Science

Keywords:

Construction Industry; Big Data; Data Analytics; Data mining; Artificial Neural Network; ANN; Compressive Strength; Environmentally Friendly Concrete; Soil Erosion; Highway Slope; Predictive Modeling; Comparative Analysis

Crossen, Samantha LokelaniInvestigation of Variability in Cognitive State Assessment based on Electroencephalogram-derived Features
Master of Science in Engineering (MSEgr), Wright State University, 2011, Biomedical Engineering
To implement adaptive aiding in modern aviation systems there is a need for accurate and reliable classification of cognitive workload. Using electroencephalogram (EEG)-derived features, it has been reported that an Artificial Neural Network (ANN) can achieve 95% or higher classification accuracy on the same day for an individual operator, but only 70% or less on a different day. To gain a further insight into this discrepancy, data from a previous study was utilized to study the classification variability. The EEG-derived features were first calculated by spectral power estimation. The variability was then analyzed by performing cognitive workload classification in which different methods of training and testing were used and different classifiers were implemented to compare classification accuracies. The classifiers include an ANN, Adaboost Algorithm, and a t-test method. The results show that when the ANN or Adaboost method is used, the amount of overlapping among training and testing data impacts the classification accuracy significantly. When there is no overlap, all classifiers can only achieve an accuracy of about 70%, with the Adaboost outperforming other classifiers slightly. By allowing some overlap, the accuracy of the ANN or Adaboost method increases significantly. It was concluded that the main source of the classification variability is the inherent variability of the EEG-derived features.

Committee:

Ping He, PhD (Advisor); James Christensen, PhD (Committee Member); Yan Liu, PhD (Committee Member)

Subjects:

Biomedical Research

Keywords:

Electroencephalogram (EEG); Artificial Neural Network (ANN); AdaBoost Algorithm; Workload Classification; Feature Variability

Dravenstott, Ronald W.Restaurant Industry Stock Price Forecasting Model Utilizing Artificial Neural Networks to Combine Fundamental and Technical Analysis
Master of Science (MS), Ohio University, 2012, Industrial and Systems Engineering (Engineering and Technology)
Stock price forecasting is a classic problem facing analysts. Forcasting models have been developed for predicting individual stocks and stock indices around the world and in numerous industries. According to a literature review, these models have yet to be applied to the restaurant industry. Strategies for forecasting typically include fundamental and technical variables. In this thesis, fundamental and technical inputs were combined into an Artificial Neural Network stock prediction model for the restaurant industry. Models were designed to forecast 1 week, 4 weeks, and 13 weeks into the future. The model performed better than the benchmarks. The prediction accuracy of the model reached as high as 60%. The model with the most success was a Multilayer Perceptron Artificial Neural Network with 2 hidden layers having 40 and 20 processing elements in those layers using the hyperbolic tangent transfer function and Delta Bar Delta learning algorithm.

Committee:

Gary Weckman, PhD (Committee Chair); Tao Yuan, PhD (Committee Member); Namkyu Park, PhD (Committee Member); Andy Snow, PhD (Committee Member)

Subjects:

Engineering; Finance; Industrial Engineering

Keywords:

Artificial Neural Network; ANN; Restaurant; Stock Price Forecast; Fundamental Analysis; Technical Analysis

Iacianci, Bryon C.Confidence Intervals on Cost Estimates When Using a Feature-based Approach
Master of Science (MS), Ohio University, 2012, Industrial and Systems Engineering (Engineering and Technology)

This research explains the methodology for deriving the confidence interval on the cost estimate of a part, when a feature-based approach is used. The components of a steam turbine are used in order to demonstrate the methodology. With a parametric approach to estimate cost, developing a confidence interval is straightforward because there is one cost-estimating relationship (CER) that incorporates a design's parameters. However, in feature-based cost estimating, there are multiple CERs that each estimate the cost of a part feature and the feature estimates are accumulated to get the total manufacturing cost. This makes deriving a confidence interval more complex, since the variance in each CER must be incorporated into determining the overall variance of the estimate.

Confidence intervals are derived for multiple CER generation techniques that utilize both regression and Artificial Neural Networks. The differences between their parametric and feature-based results are statistically tested to determine whether a difference exists. The testing shows that in 7 out 8 instances tested, the differences between the two approaches were not found to be statistically significantly different. Feature-based models are more transparent than multivariate models because exactly how each parameter affects an estimate can be easily determined. When there is no difference between the two methods than the feature-based method should be used by the analyst.

Committee:

Dale Masel, PhD (Committee Chair); Robert Judd, PhD (Committee Member); Gary Weckman, PhD (Committee Member); Gary Coombs (Committee Member)

Subjects:

Engineering; Industrial Engineering

Keywords:

cost estimating relationship; artificial neural network; confidence interval; cost prediction; CER; ANN; composite confidence interval; feature based

Selvaraj, PooraniGroup Method of Data Handling – How Does it Measure Up?
Master of Science (MS), Ohio University, 2016, Industrial and Systems Engineering (Engineering and Technology)
This study investigates the potential of Group Method of Data Handling (GMDH) as a tool for data analysis. It is a comparative study of the performance of GMDH, as opposed to other static and dynamic neural networks like Multilayer Perceptron (MLP) and Time-Lag Recurrent Neural networks (TLRN). This study focuses on the three main applications of data mining; namely, regression, classification, and time series analysis. The choice of MLP or TLRN depends on the application for which the technique was intended. At least thirteen different configurations of these networks have been built by varying a number of parameters such as the number of processing elements, learning algorithm, transfer function and types of memory (in the case of TLRN). A total of three datasets have been used for each application, which have been classified as simple, complex, and highly complex (based on coefficient of determination or proportion correct values) and small medium and large (based on either the size of the data set). Basic statistics have been used for the purpose of this classification. Statistically, GMDH performs comparatively well to MLP for regression and classification analysis. However, certain limiting factors, other than performance efficiency, fail to qualify the method for comparison with TLRN for time series analysis.

Committee:

Gary Weckman (Advisor)

Subjects:

Engineering; Industrial Engineering

Keywords:

GMDH; MLP; ANN

Lynch, Dustin ShaneAsset Allocation Technique for a Diversified Investment Portfolio Using Artificial Neural Networks
Master of Science (MS), Ohio University, 2015, Industrial and Systems Engineering (Engineering and Technology)
As part of planning for the future and retirement, people typically build their investment portfolio. Investment portfolios are made up of four different asset classes, and typically managed by one of the major investment firms such as the Edward Jones Company. This research works with artificial neural networks (ANN) and closely with an advisor from the Edward Jones Company to provide a machine learning decision making aid for them to use when allocating the four main asset classes that make up a portfolio. The asset class prediction results and trends are then compared by the advisors consulted to decide if this methodology would be a useful aid during high volatility times in the stock market, such as the market crash of 2008. The use of this successful machine learning aid will benefit the investment portfolio that shows promise for yielding higher return on investment (ROI). This research was determined to be a successful machine learning aid to assist advisors with the asset allocation of an investment portfolio.

Committee:

Gary Weckman, Ph.D. (Advisor); Andy Snow, Ph.D. (Committee Member); Tao Yuan, Ph.D. (Committee Member); Namkyu Park, Ph.D. (Committee Member)

Subjects:

Engineering; Finance

Keywords:

Asset Allocation; Artificial Neural Networks; ANN; Financial Engineering; Investment Portfolio Prediction

Ko, YoungseoNonlinear Device Measurement, Characterization, and Modeling for High Power RF Applications
Doctor of Philosophy, The Ohio State University, 2013, Electrical and Computer Engineering
To satisfy power and frequency requirements for commercial RF applications, various semiconductor materials and device structures have been introduced and developed rapidly. These semiconductors need to handle modulated signals exhibiting high peak to average power ratio (PAPR) and wide modulation bandwidth to meet the requirements of modern wireless standards. In this context, measurements with continuous wave (CW) single tone or two-tone excitations are no longer sufficient to characterize power amplifiers (PAs) for such applications due to the inherent memory effects present in PAs. In this thesis a new testbed is developed with a Large Signal Network Analyzer (LSNA) to perform multi-harmonic broadband measurements for periodically modulated signals with bandwidth exceeding the IF receiver bandwidth of the LSNA. Also a new pulsed real-time active load-pull (RTALP) measurement system which is developed and used to characterize devices under iso-memory effect is reported in this thesis. With the LSNA system, the multiple recording technique are combined in order to reduce the long measurement time and bypass the pulse desensitization problem. With this novel combined technique, output power and gain at 1 dB compression gain plane are extracted from a double power-sweep measurements of the input and output power levels. It is necessary for successful device modeling to account for memory effects such as self-heating, traps in HEMTs and parasitic bipolar junction transistor in FETs. Due to the slow time response of these memory effects, fast pulsed-IV/pulsed- RF measurements have been used to maintain constant device temperature, body BJT voltage in FETs or trap state in HEMTs. Therefore, new microwave characterization techniques are also presented to characterize these unwanted memory effects by estimating the internal device operating temperature and extracting the trap relaxation time-constants in AlGaN/GaN HEMTs. Having characterized results under the developedmeasurement system, a new measurement based model with artificial neuron network (ANN) is finally proposed and applied to a SOI MOSFET device. The verification results demonstrate that the time-consuming measurement process can be dramatically reduced by using RTALP measurement data, and that fairly accurate large-signal RF device model can be easily extracted from these measurements using the ANN approach.

Committee:

Patrick Roblin (Advisor); Waleed Khalil (Committee Member); Roberto Rojas-Teran (Committee Member)

Subjects:

Electrical Engineering

Keywords:

ANN; MOSFET; RTALP; LSNA; Device Modeling.

Avila, Beth Eileen“I Would Prevent You from Further Violence”: Women, Pirates, and the Problem of Violence in the Antebellum American Imagination
Doctor of Philosophy, The Ohio State University, 2016, English
“'I Would Prevent You from Further Violence': Women, Pirates, and the Problem of Violence in the Antebellum American Imagination" analyzes how antebellum American pirate stories used the figure of the pirate to explore the problem of violence and the role women play in opposing violent men. This project joins ongoing conversations about women in the nineteenth century in which scholars, such as Nina Baym, Mary Kelley, and Mary Ryan, have made key contributions by recovering a domestic model of nineteenth-century womanhood. As my work demonstrates, antebellum Americans were similarly invested in a more adventurous, and sometimes violent, model of womanhood that was built upon the figure of the gentleman pirate and placed in opposition to violent men. I argue that it is important to think about the pirate story and the figure of the pirate, not only in the context in which it has come to be known—escapist fantasies written for boys and young men—but as a place where authors reinforced, modified, and established different models of gender roles. Frequently within the mid-nineteenth-century American pirate story, authors answered the question of who is allowed to be violent by demonstrating that women had the capacity for violence and constructing scenarios illustrating that women were often the only ones in a position to forcibly oppose violent men. The pirate story uniquely blends different narrative conventions: adventure stories that are often believed to appeal to male audiences and domestic scenarios that are usually understood to resonate with female readers. Although historical and fictional pirates of other eras and geographical locations have been examined, little scholarship has focused on piracy in the antebellum American imagination, even though the figure of the pirate continued to proliferate, especially in popular fiction, throughout the nineteenth century. My project addresses this gap not only by demonstrating the importance of pirates in nineteenth-century American fiction, but also by exploring how American authors, such as James Fenimore Cooper, Catherine Maria Sedgwick, J. H. Ingraham, and Eliza Ann Dupuy, were responding to and revising earlier British depictions of the figure of the pirate by Byron and Walter Scott. In addition to its unique focus on pirates in antebellum American fiction, my project explores a body of popular texts that has been neglected by scholars. These ephemeral stories are difficult to obtain outside of archives and libraries; however, they were extremely popular in their own time. While scholarship has shifted to recognize the value in previously dismissed popular texts, story papers and shilling novelettes have not yet been thoroughly analyzed, and my project seeks to reintroduce a small portion of these texts through archival research centering on the work of Maturin Murray Ballou and Benjamin Barker. By placing transatlantic canonical texts in conversation with formerly neglected ephemeral mediums, I explore stories that resonated with audiences and proposed unconventional violent and heroic models of womanhood, thus disrupting the idea that there was a monolithic version of womanhood in antebellum America.

Committee:

Sara Crosby (Advisor); Andrea Williams (Committee Member); Susan Williams (Committee Member)

Subjects:

American History; American Literature; American Studies; British and Irish Literature; Gender; Literature; Womens Studies

Keywords:

pirates; women; men; gender roles; violence; genre; antebellum; nineteenth century; United States; transatlantic; story papers; James Fenimore Cooper; Catharine Maria Sedgwick; Joseph Holt Ingraham; Eliza Ann Dupuy; Maturin Murray Ballou; Benjamin Barker

Thomas, Philip S.A Reinforcement Learning Controller for Functional Electrical Stimulation of a Human Arm
Master of Sciences, Case Western Reserve University, 2009, EECS - Computer and Information Sciences
This thesis demonstrates the feasibility of using reinforcement learning (RL) for functional electrical stimulation (FES) control of a human arm as an improvement over (i) previous closed-loop controllers for upper extremities that are unable to adapt to changing system dynamics and (ii) previous RL controllers that required thousands of arm movements to learn. We describe the relevance of the control task and how it can be applied to help people with spinal cord injuries. We also provide simulations that show previous closed-loop controllers are insufficient. We provide background on possible RL techniques for control, focusing on a continuous actor-critic architecture that uses function approximators for its mappings. We test various function approximators, including Artificial Neural Networks (ANNs) and Locally Weighted Regression (LWR) for this purpose. Next, we introduce a novel function approximator, Incremental Locally Weighted Regression (ILWR), which is particularly suited for use in our RL architecture. We then design, implement, and perform clinically relevant tests using ANNs for the two mappings in the continuous actor-critic. During these trials, unexpected behavior is observed and eventually used to create a hybrid controller (that switches among different learning parameter sets) that can both adapt to changes in arm dynamics in 200 to 300 arm movements and remain stable in the long-term. A non-switching controller with similar performance is achieved using ILWR in place of an ANN for the controller's critic mapping.

Committee:

Michael Branicky, PhD (Advisor); Antonie van den Bogert, PhD (Committee Member); Soumya Ray, PhD (Committee Member)

Subjects:

Computer Science

Keywords:

reinforcement learning; locally wieghted regression; incremental locally weighted regression; LWR; ILWR; ANN; actor-critic; continuous actor-critic; functional electrical stimulation; FES; adaptive control

Wikle, Olivia MarieMortal Sounds and Sacred Strains: Ann Radcliffe's Incorporation of Music in The Mysteries of Udolpho
Master of Arts, The Ohio State University, 2016, Music
The British gothic novel arose during the second half of the eighteenth century, amidst the effects of the Enlightenment and the beginning of Romanticism. The gothic genre drew on a conception of a superstitious past, depicting marvelous and otherworldly subject matter in order to express growing feelings of anxiety caused by urbanization, revolution, and a displacement of religion. In evoking these anxieties, gothic writing also drew on commonly held aesthetic theories of the eighteenth century, most notably the theory of the sublime. In order to enhance the experience of sublime emotions in her readers, the gothic novelist Ann Radcliffe (1764-1823) incorporated descriptions of sound and music in her novels, particularly during apparently supernatural scenes. Radcliffe was the first gothic author to implement music and sound in this way, and her sonic descriptions had profound effects on the writing of gothic authors that came after her. Radcliffe’s descriptions of sound and music in her 1794 novel, The Mysteries of Udolpho, bear similarities with the mimetic techniques of eighteenth-century art music that was being performed in London during Radcliffe’s lifetime. Biographical information indicates that Radcliffe was fond of attending musical performances in late eighteenth-century London, and was familiar with mimetic musical techniques that were understood by aesthetic theorists to evoke sublime emotions. Two compositions in particular, Antonio Sacchini’s Armida and George Frederic Handel’s Israel in Egypt, display musical characteristics that are similar to those that Radcliffe describes. These similarities indicate that her writing was likely influenced by the performances she heard in late eighteenth-century Britain.

Committee:

Danielle Fosler-Lussier (Advisor); Ryan Skinner (Committee Member); Clare Simmons (Committee Member)

Subjects:

British and Irish Literature; Music

Keywords:

Gothic; Music; Ann Radcliffe; Eighteenth-Century Music

PARTRIDGE, LAURA ALLISON“Far From Silenced”: The Altered Books of Ann Hamilton, 1991–1994
MA, University of Cincinnati, 2007, Design, Architecture, Art and Planning : Art History
This study offers an in-depth analysis of the “altered books” in three installations by Ann Hamilton (b. 1956 Lima, Ohio) from the early 1990s. Hamilton’s installations are intended to establish a sensory experience that responds to the sites in which she creates them. She often includes a person who tends the installation, that is, a performer who is called a “tender,” who acts unaware of the presence of spectators, and who is wholly engrossed in the task with which s/he has been charged. The tenders in the works upon which I will focus sat at a table systematically removing lines of text from books, that is, they altered the books. In my analysis I will discuss all of the components within each installation, focusing on the altered books and how they played a central role in securing language as a major theme in these works.

Committee:

Kimberly Paice (Advisor)

Subjects:

Art History

Keywords:

Ann Hamilton; installations; altered books; language

Young, Ryan F.Utilization of a Neural Network to Improve Fuel Maps of an Air-Cooled Internal Combustion Engine
Master of Science (MS), Ohio University, 2010, Industrial and Systems Engineering (Engineering and Technology)
Fuel maps are utilized by the fuel injection system as a guide for accurate delivery of fuel under a specified load. A fuel map is determined by the manufacturer and usually not manipulated. This research involves exhaust gas oxygen data collection using an original equipment engine control module (ECM), artificial neural network (ANN) modeling, response surface generation that will act as the new fuel map, implementing the map into the ECM, and testing. ANN modeling is used first to predict volumetric efficiency (VE) values in the fuel map, then used to optimize the VE values based on the air to fuel ratio. The results are then compared with an alternative optimization technique and the original equipment fuel map. Optimization of the fuel map will provide physical performance, economic, and environmental gains. Applying this methodology would allow the fuel map to be updated using little expert knowledge.

Committee:

Gary R. Weckman, PhD (Advisor); Helmut Paschold, PhD (Committee Member); Namkyu Park, PhD (Committee Member); Tao Yuan, PhD (Committee Member)

Subjects:

Artificial Intelligence; Industrial Engineering; Systems Design; Transportation

Keywords:

Optimize Fuel Map; ANN; ECM; Internal Combustion Engine; O2 sensor

Wilmot, Timothy AllenIntelligent Controls for a Semi-Active Hydraulic Prosthetic Knee
Master of Science in Electrical Engineering, Cleveland State University, 2011, Fenn College of Engineering
We discuss open loop control development and simulation results for a semi-active above-knee prosthesis. The control signal consists of two hydraulic valve settings. These valves control a rotary actuator that provides torque to the prosthetic knee. We develop open loop control using biogeography-based optimization (BBO), which is a recently developed evolutionary algorithm, and gradient descent. We use gradient descent to show that the control generated by BBO is locally optimal. This research contributes to the field of evolutionary algorithms by demonstrating that BBO is successful at finding optimal solutions to complex, real-world, nonlinear, time varying control problems. The research contributes to the field of prosthetics by showing that it is possible to find effective open loop control signals for a newly proposed semi-active hydraulic knee prosthesis. The control algorithm provides knee angle tracking with an RMS error of 7.9 degrees, and thigh angle tracking with an RMS error of 4.7 degrees. Robustness tests show that the BBO control solution is affected very little by disturbances added during the simulation. However, the open loop control is very sensitive to the initial conditions. So a closed loop control is needed to mitigate the effects of varying initial conditions. We implement a proportional, integral, derivative (PID) controller for the prosthesis and show that it is not a sufficient form of closed loop control. Instead, we implement artificial neural networks (ANNs) as the mechanism for closed loop control. We show that ANNs can greatly improve performance when noise and disturbance cause high tracking errors, thus reducing the risk of stumbles and falls. We also show that ANNs are able to improve average performance by as much as 8% over open loop control. We also discuss embedded system implementation with a microcontroller and associated hardware and software.

Committee:

Dan Simon, PhD (Advisor); Fuquin Xiong, PhD (Committee Member); Lili Dong, PhD (Committee Member)

Subjects:

Electrical Engineering; Engineering

Keywords:

prosthetic control; ANN; artificial neural network; BBO; biogeography based optimization; intelligent control; nonlinear control problem; time varying control problem; evolutionary algorithm; gradient descent

Yarlagadda, ManideepAn analysis of NOx and PM emissions in idling and moving conditions of buses with EGR and Non-EGR engines running on biodiesel
Master of Science, University of Toledo, 2016, Civil Engineering
Biodiesel is an alternate to diesel for transit buses due to its environmental benefits. However, NOx and particulate matter emissions may be an issue in the use of biodiesel. The major objective of this experimental thesis was to study tail pipe emissions from transit buses during daily routine operations. This thesis focuses on the trends of NOx and particulate matter emissions collected from buses with EGR and NON-EGR engines during their total run times. To further categorize and elaborate our findings, the run time was divided into both idling and running conditions. In order to achieve comprehensive results, the idling and running conditions were further segregated into two different cases, i.e., cold idling and hot idling conditions. The running conditions were divided into acceleration, deceleration, motion in variable speeds and partial idle modes. The NOx emission values were collected and analyzed for all the conditions and modes described above. The particulate matter emissions were collected and analyzed in idle conditions. It was learned that hotter engines produced lower emissions when compared to cold engine conditions. The experiments and analysis of NOx emissions concluded that maximum emissions were found in the acceleration condition. A Mexa-720 Horiba NOx analyzer was used to measure NOx emissions and Cummins in-site 6 equipment and software program were used for engine data collection during the field study. The experiments were carried out on both transit buses with EGR and NON-EGR engines. The particulate matter emissions collection was carried out with quartz filter papers and a CATCH CAN instrument. An EDS X-Max 50mm2 / FEI Quanta 3D FEG Dual Beam Electron Microscope was used for the EDS analysis of PM emissions and the ICP-MS was carried out using Xseries 2. The transit buses are used by Toledo Area Regional Transit Authority (TARTA). Both the buses were fueled with B5 grade biodiesel without making any engine modifications and the study was conducted during the summer and fall of 2015. The emission values were collected along with the consideration of various engine parameters such as engine temperature, exhaust gas pressure, fuel flow rate command, diesel oxidation catalyst intake temperature and the diesel particulate filter intake temperature. The collected NOx emission values were analyzed, as a function of time, with the help of three different regression techniques and obtained the best results with the Random Forest Regression algorithm. A NOx emission prediction model was established as a function of the engine parameters using the field data and regression results. Elemental analysis was performed on the particulate matter emissions and it was concluded that trace metal and carbon concentrations were higher in the NON-EGR engine buses in comparison to the EGR engine buses.

Committee:

Dr. ASHOK KUMAR (Committee Chair); Dr. DONG-SHIK KIM (Committee Co-Chair); Dr. LIANGBO HU (Committee Member)

Subjects:

Chemical Engineering; Civil Engineering; Environmental Engineering

Keywords:

NOx, EGR, ANN, PM, ICP-MS, EDS-SEM

Khandelwal, Aashish S.Evaluating Grey-box Models in Highly and Slightly Correlated Imbalanced Data Sets
Master of Science (MS), Ohio University, 2010, Industrial and Systems Engineering (Engineering and Technology)
Data analysis has been one of the most challenging subjects for researchers in any field. Understanding the data in the right way has been the biggest challenge for any data-analyst. Imbalance and correlation are the properties that prevent researchers from constructing good non-linear models. A combination of white-box model (first principles and surface equations) and black-box model (ANN and regression) is used to construct a grey-box model, which has performed better than former models individually. Performance of grey-box models with four different databases is analyzed. An attempt has also been made to find a better way of constructing a grey-box and compare its performance with multi linear regression.

Committee:

Gary Weckman, R (Advisor)

Subjects:

Industrial Engineering

Keywords:

Grey-box; ANN; Regression

Owen, Kate Marie NovotnyModes of the Flesh: A Poetics of Literary Embodiment in the Long Eighteenth Century
Doctor of Philosophy, The Ohio State University, 2017, English
Modes of the Flesh considers the ways that literary form—mode, in particular—shapes the representation of the human body in British literature from approximately 1660-1800. Focusing on the allegorical, satirical, pornographic, and gothic modes, this project aims to expand our conception of literary embodiment, establish the represented body as a formal element, and make embodiment central to our understanding of the textual representation of human beings. Because modally-inflected literary bodies engage the same kinds of ontological and epistemological questions entertained by this period’s empiricist philosophy, I argue that mode offers its own kind of philosophy of the body. But, because modal bodies engage these questions with a very different set of tools, the results are often provocatively at odds with mainstream philosophical discourse. Existing scholarship on the literary body tends either to analyze the way a body is represented in order to better understand the work’s themes or meanings, or to argue that the way a body is represented reflects historical or theoretical models of embodiment. This dissertation differs from the first tendency by offering a theory of the represented body, and therefore taking the body as an object, not an instrument, of study. It diverges from the second tendency by arguing that the way bodies are presented in literature has as much to do with the kind of text they appear in as with scientific, theological, social, or other extra-literary understandings of the body. In each chapter, I focus on a significant mode of Restoration and eighteenth-century literature, and a particular aspect of literary embodiment. The first chapter, on the allegorical mode and bodily matter, thinks about the function of materiality in a mode commonly associated with abstraction and interpretation. The second chapter, which considers the satirical mode and bodily form, explores the role of abstract form in satirical conceptions of personhood and in the ordering of the satirical universe. The third chapter focuses on the pornographic mode and the body as sensation in order to re-think the mechanics and ethics of pornography. And, in the fourth chapter, I consider the gothic mode in terms of the two kinds of embodiment it pits against one another (the body as figure, and the body as subjectivity) in order to consider its fantastical exaggerations of empirical—and especially skeptical—epistemology. In the coda, I meditate on the desire for disembodiment and the fundamental importance of the body to the representation of human experience. In the course of developing a formal theory of the body in Restoration and eighteenth-century literature, Modes of the Flesh challenges the tendency in literary studies to marginalize the body in favor of the mind. By foregrounding the body and mode, it expands the scope of character studies and challenges the centrality of the rise of the novel to the field of eighteenth-century studies.

Committee:

Sandra Macpherson (Advisor); David Brewer (Committee Member); Robyn Warhol (Committee Member)

Subjects:

British and Irish Literature; Literature

Keywords:

body; mode; character; form; allegory; satire; pornography; gothic; eighteenth century; Restoration; empiricism; poetics; theory; representation; Jonathan Swift; John Bunyan; John Milton; John Cleland; Ann Radcliffe; Jane Austen; Earl of Rochester

Taylor, Brent S.Utilizing ANNs to Improve the Forecast for Tire Demand
Master of Science (MS), Ohio University, 2015, Industrial and Systems Engineering (Engineering and Technology)
This study is an initial attempt to investigate the relationship between economic factors and monthly tire sales, using artificial neural networks (ANNs) and comparing the results to stepwise regression. Data for this research were collected through a privately held tire warehouse located in Wheeling, West Virginia. Research has shown that artificial neural network models have been successfully applied to many real world forecasting applications. However, up to this date no research has been found using artificial neural networks and economic factors to predict tire demand. The first part of this research describes why the chosen economic factors were selected for this study and explains the initial methodology with results. The next stage of the research gives details on why the methodology was revised and also clarifies why Google Trends and additional mathematical inputs were applied to the study. The final research focused on separating the master database into three different categories based on selling percentages. The results of the study show that the artificial neural network models were capable of forecasting the number of high selling tires, with a validation technique, but were unable to be applied sufficiently for the medium and low selling products.

Committee:

Gary Weckman, Ph.D. (Advisor)

Subjects:

Engineering; Industrial Engineering

Keywords:

Artificial Neural Networks; Tire Forecasting; Demand Planning; ANN; Tire Demand; Economic Factors

Littlejohn, AmonteHOPEFUL HOSTILITY:AN ANALYSIS OF THE EVOLUTION OF AMERICAN NATURALISM
Master of Arts in English, Cleveland State University, 2011, College of Liberal Arts and Social Sciences
American Naturalism has a reputation of being a reductive and often times violent genre, but in its brutality exists a lens to examine adverse social conditions and practices of modern and historical society. Evolved from its precursor in European Naturalism, American Naturalism would undergo adaptations to make the genre more relevant to the American audience, authors like Frank Norris and Stephen Crane each tailoring their naturalistic novels to cater to their respective times. Since then, the genre has gone as a style that is as difficult to define as it is to accept, American Naturalism receiving criticisms and detractions with each novel written. Nevertheless, the genre has endured and only further adapted with America’s constantly changing social climate. To assess and examine the adaptations in American Naturalism, texts written long-after American Naturalism’s inception were analyzed through Valerie Smith’s theory of intersectionality. Rather than focusing on one particular aspect of a text, Smith’s intersectionality examines multiple components in a subject and examines not only their individual roles but their relationship with one another as well. The novels chosen, Ann Petry’s The Street (1941) and Max Brooks’ World War Z (2006), are first qualified as American Naturalistic texts by way of genre hallmarks before Smith’s theory is applied to show not only how the hallmarks contribute to the novels individually, but how those same identifiers have evolved over time. This thesis focuses primarily on the evolutions in American Naturalism’s narratological method and its expansion of the naturalistic conclusion.

Committee:

Adrienne Gosselin, PhD (Advisor); David (Ted) Lardner, PhD (Committee Member); Stella Singer, PhD (Committee Member)

Subjects:

Literature

Keywords:

American Naturalism; horror fiction; zombies; naturalism; narratology; violence; black fiction; max brooks; world war z; ann petry; the street; criticism

Anderson, Jerone S.A Study of Nutrient Dynamics in Old Woman Creek Using Artificial Neural Networks and Bayesian Belief Networks
Master of Science (MS), Ohio University, 2009, Industrial and Systems Engineering (Engineering and Technology)
The Old Woman Creek National Estuary is studied in this project to evaluate effective modelling techniques for predicting Net Ecosystem Metabolism (NEM). NEM is modelled using artificial neural networks, Bayesian belief networks, and a hybrid model. A variety of data preprocessing techniques are considered prior to model development. The effects of discretization on model development are considered and discrete data is ultimately used to produce models which classify NEM into three ranges based on inputs with information significance. Artificial neural networks are found to be the most accurate for classification while Bayesian belief networks are found to provide a better framework for dynamically predicting NEM as inputs are changed.

Committee:

Gary R. Weckman, PhD (Advisor); David Millie, PhD (Committee Member); Kevin Berisso, PhD (Committee Member); Diana Schwerha, PhD (Committee Member)

Subjects:

Ecology; Engineering; Environmental Engineering; Industrial Engineering

Keywords:

BBN; ANN; ecology; NEM; Bayesian Belief Networks; Artificial Neural Networks; computer modelling

McIntyre, Heather DawnMystical Motherhood: Blending Ecstatic Religious Experience with Feminist Discourse in Appalachian Fiction
Master of Arts (MA), Bowling Green State University, 2010, English/Literature

Appalachia is a region steeped in religious tradition. Religious discourse permeates the way Appalachians think and speak and the stories they tell; for many, it shapes their sense of Self within their families and culture, influencing their understanding of what it means to be moral, functioning members of society. Furthermore, religious discourse is often used as a rhetorical tool to help problematize specific political and cultural practices within Appalachian communities. It is not surprising, then, that author Bobbie Ann Mason weaves ecstatic and mystical religious experiences into her work as she addresses the concept of gender roles in Appalachian society. Mason’s novel Feather Crowns highlights ecstatic religious experience in its depiction of how ecstatic religious experiences and mountain faith structures can be exploited to keep women defined by rigidly-determined gender roles.

How does Mason craft ecstatic religious experience in ways that emphasize its double-edged nature as both a reification of women’s entrapment within gender norms and a tool that allows individual women’s defiance of specific gendered expectations? Furthermore, how do these works frame women’s sense of Self within their communities? How do they call into question long-held cultural stereotypes concerning Appalachia and its citizens. I utilize research into gender identity, feminism, and the positive and negative aspects of ecstatic religious experience in order to answer questions. As I demonstrate by doing a close reading of the works of Louis Althusser, Amy Hollywood, George Bataille, and Bobbie Ann Mason, literary depictions of ecstatic religious experience in Appalachia can be used to make palpable and question the religious and familial ideologies concerning women’s gendered position in society. Mason’s work, in particular, brings into dramatic light the pain, frustration, and helplessness felt by women whose ecstatic religious experience comes into conflict with familial and/or social expectations

Committee:

Erin Labbie, PhD (Committee Chair); Becca Cragin, PhD (Committee Member); Kristine Blair, PhD (Committee Member)

Subjects:

English literature

Keywords:

Appalachia; ecstasy; feminism; gender; religion; Bobbie Ann Mason