Master of Science in Mechanical Engineering, Cleveland State University, 2023, Washkewicz College of Engineering

In structural design, the distribution of wind-induced pressure exerted on structures is crucial. The pressure distribution for a particular building is often determined by scale model tests in boundary layer wind tunnels (BLWTs). For all combinations of interesting building shapes and wind factors, experiments with BLWTs must be done. Resource or physical testing restrictions may limit the acquisition of needed data because this procedure might be time- and resource-intensive. Finding a trustworthy method to cyber-enhance data-collecting operations in BLWTs is therefore sought. This research analyzes how machine learning approaches may improve traditional BLWT modeling to increase the information obtained from tests while proportionally lowering the work needed to complete them. The more general question centers on how a machine learning-enhanced method ultimately leads to approaches that learn as data are collected and subsequently optimize the execution of experiments to shorten the time needed to complete user-specified objectives. 3 Different Machine Learning models, namely, Support vector regressors, Gradient Boosting regressors, and Feed Forward Neural networks were used to predict the surface Averaged Mean pressure coefficients cp on Isolated high-rise buildings. The models were trained to predict average cp for missing angles and also used to train for varying dimensions. Both global and local approaches to training the models were used and compared. The Tokyo Polytechnic University's Aerodynamic Database for Isolated High-rise buildings was used to train all the models in this study. Local and global prediction approaches were used for the DNN and GBRT models and no considerable difference has been found between them. The DNN model showed the best accuracy with (R2 > 99%, MSE < 1.5%) among the used models for both missing angles and missing dimensions, and the other two models also showed high accuracy with (R2 > 97%, MSE < 4%).

Committee: Navid Goudarzi (Committee Chair); Prabaha Sikder (Committee Member); Mustafa Usta (Committee Member) Subjects: Artificial Intelligence; Design; Engineering; Urban Planning

Master of Science, Miami University, 2021, Computational Science and Engineering

As power electronics permeate critical infrastructure in modern society, more precise and effective diagnostic methods are required to improve system reliability as well as reduce maintenance costs and unexpected failures. Prognostic and Health Management (PHM) systems are real-time analysis hardware that estimate device health by monitoring underlying failure mechanisms. While several variants of PHM methods have been explored, the use of electromagnetic interference (EMI) as a conditional monitoring tool, referred to as E-PHM, has received limited attention despite its utility as a sensitive and non-invasive prognostic tool. This research demonstrates the feasibility of E-PHM techniques to measure, in real-time, the junction temperature of power devices using machine learning algorithms (MLAs). This is accomplished, in situ, without interruption of device operation and without altering the system's performance. Semiconductor operating parameters are sensitive to changes in temperature, altering device behavior. These changes in behavior are reflected in the electromagnetic spectrum of the circuit. Preliminary research has classified changes in EMI via Support Vector Machine algorithm to predict device junction temperature. The proposed approach will shift from classification-based models, such as the SVM, to regression-based models to improve accuracy and precision in junction temperature prediction.

Committee: Mark Scott (Advisor); Miao Wang (Committee Member); Chi-Hao Cheng (Committee Member) Subjects: Electrical Engineering; Engineering

Master of Science (MS), Wright State University, 2018, Computer Science

Sickle Cell Disease (SCD) is a hereditary disorder in red blood cells that can lead to excruciating pain episodes. SCD causes the normal red blood cells to distort its shape and turn into sickle shape. The distorted shape makes the hemoglobin inflexible and stick to the walls of the vessels thereby obstructing the free flow of blood and eventually making the tissues suffer from lack of oxygen. The lack of oxygen causes serious problems including Acute Chest Syndrome (ACS), stroke, infection, organ damage, and over the lifetime an SCD can harm a persons spleen, brain, kidneys, eyes, bones. Sickling of RBC can be triggered by a number of conditions such as dehydration, acidity, low levels of oxygen, stress, and change in temperature. There is no specific medication for pain crisis and the signs and symptoms varies from person to person, making it difficult to provide a common treatment for SCD and understanding the disease. It is believed that 90,000 to 100,000 American are affected by SCD. Myriad number of studies have been working on gaining better understanding of the disease and predict pain crisis and pain level. These studies help people to mitigate or prevent pain crisis by taking precautions. However, no study has used clinical notes to predict pain score and pain sentiment. Clinical notes provide patient specific information including procedures and medication; and can therefore help in predicting accurate scores. Our study focuses on four research problems namely patient informative, pain informactive, pain sentiment and pain scores using SCD data. Notes are taken for a patient during hospitalization but only few provide beneficial information, therefore patient informative and pain informative helps healthcare professionals to scan through the notes that can pro- vide valuable information from all the clinical notes maintained. Pain sentiment and pain score predict the change in pain and pain level for a particular note. Our study experimented with two fea (open full item for complete abstract)

Committee: Tanvi Banerjee Ph.D. (Advisor); Michelle Cheatham Ph.D. (Committee Member); Mateen Rizki Ph.D. (Committee Member) Subjects: Computer Science

Doctor of Philosophy, The Ohio State University, 2017, Food, Agricultural and Biological Engineering

The objectives of this research were to: (i) investigate different aspects of long-term trend detection in monthly, seasonal, and annual streamflow; (ii) assess climate change impacts on the upper Senegal River basin stream-flow; and (iii) evaluate data driven models to forecast daily stream at the upper Senegal River basin. A preliminary study investigated long-term trends on stream-flow at the upper Senegal River basin. Results showed a trend of decreasing annual stream flow; but this was not significant at p < 0.05 for the whole period. However, when integrating various temporal breaking points, a decreasing trend was significant before the first breaking point (1976) but was reversed for the period of 1976 to 1993. For the monthly series, all months exhibit a non-significant decreasing trend except for the month of June that had an increasing trend. The seasonal series showed a decreasing trend that was significant (p <0.05) at MAMJ season. The extremely low and high daily streamflow had significant positive and negative trends, respectively. This research used five General Circulation Models (CNRM, CSIRO, HadGEM2-CC, HadGEM2-ES, and MIROC5), two RCP (RCP4.5 and RCP8.5) scenarios, and the GR4J hydrological model to evaluate the impact of climate change on stream flow in the near future. The results showed that for all models, there were simulated increases in mean monthly temperature under the RCP 4.5 and RCP 8.5. Increases of temperature ranged from 0.54° C to 2.32° C under RCP 4.5 scenario and from 1.12° C to 2.78° C under RCP8.5. However, models were not consistently in agreement in the direction and magnitude of future precipitation changes for monthly rainfall. Some models predicted an increase while other a decrease. The multi-model ensemble projected a decrease of rainfall for all months except for September. The greatest precipitation increases were in September, at 15.6 mm and 10.1 mm with RCP4.5 and RCP8.5, respectively. The greatest precipitation d (open full item for complete abstract)

Committee: Larry C Brown PhD (Advisor) Subjects: Agricultural Engineering

Master of Science, The Ohio State University, 2017, Public Health

Introduction: Identifying the key elements associated with hospital readmission is critical in terms of controlling the cost for hospitals and improving the care quality for patients. Our goal is to compare three different statistical models of predicting readmission rate in pediatric trauma patients and identify important risk factors. Methods: Logistic regression, random forest and support vector machine are popular statistical models for predicting binary outcomes. We apply these three methods to the Healthcare Cost and Utilization Project (HCUP) National Readmissions Database (NRD) 2013-2014 to compare their predictive performance for readmission. Results: The Support Vector Machine method with linear function has the greatest mean AUC (0.6724) across 10-fold cross validation in the training set. The logistic regression model has the greatest AUC value (0.6862) in the validation set. Support Vector Machine with linear function (AUC=0.6842) has the lowest misclassification rate and highest sensitivity in the validation set. Conclusions: Pediatric trauma patients have a low readmission risk. The key factors of readmission are CCS diagnosis, age and mechanism of trauma.

Committee: Bo Lu (Advisor); Chi Song (Committee Member) Subjects: Biostatistics

Master of Science in Engineering, University of Toledo, 2013, Engineering (Computer Science)

Radon-222 and its parent Radium-226 are naturally occurring radioactive decay products of Uranium-238. The US Environmental Protection Agency (USEPA) attributes about 10 percent of lung cancer cases that is `around 21,000 deaths per year' in the United States, caused due to indoor radon. The USEPA has categorized Ohio as a Zone 1 state (i.e. the average indoor radon screening level greater than 4 picocuries per liter). In order to implement preventive measures, it is necessary to know radon concentration levels in all the zip codes of a geographic area. However, it is not possible to survey all the zip codes, owing to reasons such as inapproachability. In such places where radon data are unavailable, several interpolation techniques are used to estimate the radon concentrations. This thesis presents a comparison between recently developed interpolation techniques to new techniques such as Support Vector Regression (SVR), and Random Forest Regression (RFR). Recently developed interpolation techniques include Artificial Neural Network (ANN), Knowledge Based Neural Networks (KBNN), Correction-Based Artificial Neural Networks (CBNN) and the conventional interpolation techniques such as Kriging, Local Polynomial Interpolation (LPI), Global Polynomial Interpolation (GPI) and Radial Basis Function (RBF) using the K-fold cross validation method.

Committee: William Acosta (Committee Chair); Vijay Devabhaktuni (Committee Co-Chair); Ashok Kumar (Committee Member); Rob Green (Committee Member) Subjects: Computer Science

PhD, University of Cincinnati, 2007, Arts and Sciences : Physics

In this dissertation, we develop statistical and machine learning methods for problems in biological systems and processes. In particular, we are interested in two problems–predicting structural properties for membrane proteins and clustering genes based on microarray experiments. In the membrane protein problem, we introduce a compact representation for amino acids, and build a neural network predictor based on it to identify transmembrane domains for membrane proteins. Membrane proteins are divided into two classes based on the secondary structure of the parts spanning the bilayer lipids: alpha-helical and beta-barrel membrane proteins. We further build a support regression model to predict the lipid exposed levels for the amino acids within the transmembrane domains in alpha-helical membrane proteins. We also develop methods to predict pore-forming residues for beta-barrel membrane proteins. In the other problem, we apply a context-specific Bayesian clustering model to cluster genes based on their expression levels and cDNA copy numbers. This dissertation is organized as follows. Chapter 1 introduces the most relevant biology and statistical and machine learning methods. Chapters 2 and 3 focus on prediction of transmembrane domains for the alpha-helix and the beta-barrel, respectively. Chapter 4 discusses the prediction of relative lipid accessibility, a different structural property for membrane proteins. The final chapter addresses the gene clustering approach.

Committee: Dr. Mark Jarrell (Advisor) Subjects: Physics, Molecular

Doctor of Philosophy, The Ohio State University, 2010, Physics

Metal oxides are a staple of the sensor industry. The combination of their sensitivity to a number of gases, and the electrical nature of their sensing mechanism, make the particularly attractive in solid state devices. The high temperature stability of the ceramic material also make them ideal for detecting combustion byproducts where exhaust temperatures can be high. However, problems do exist with metal oxide sensors. They are not very selective as they all tend to be sensitive to a number of reduction and oxidation reactions on the oxide's surface. This makes sensors with large numbers of sensors interesting to study as a method for introducing orthogonality to the system. Also, the sensors tend to suffer from long term drift for a number of reasons. In this thesis I will develop a system for intelligently modeling metal oxide sensors and determining their suitability for use in large arrays designed to analyze exhaust gas streams. It will introduce prior knowledge of the metal oxide sensors' response mechanisms in order to produce a response function for each sensor from sparse training data. The system will use the same technique to model and remove any long term drift from the sensor response. It will also provide an efficient means for determining the orthogonality of the sensor to determine whether they are useful in gas sensing arrays. The system is based on least squares support vector regression using the reciprocal kernel. The reciprocal kernel is introduced along with a method of optimizing the free parameters of the reciprocal kernel support vector machine. The reciprocal kernel is shown to be simpler and to perform better than an earlier kernel, the modified reciprocal kernel. Least squares support vector regression is chosen as it uses all of the training points and an emphasis was placed throughout this research for extracting the maximum information from very sparse data. The reciprocal kernel is shown to be effective in modeling the sensor respon (open full item for complete abstract)

Committee: Bruce Patton PhD (Advisor); Ralf Bundschuh PhD (Committee Member); David Stroud PhD (Committee Member); Patricia Morris PhD (Committee Member); Edward Overman PhD (Committee Member) Subjects: Artificial Intelligence; Materials Science; Physics

Doctor of Philosophy, The Ohio State University, 2008, Statistics

We consider statistical procedures for feature selection defined by a family of regularizationproblems with convex piecewise linear loss functions and penalties of l1 or l∞ nature. For example, quantile regression and support vector machines with l1 norm penalty fall into the category. Computationally, the regularization problems are linear programming (LP) problems indexed by a single parameter, which are known as “parametric cost LP” or “parametric right-hand-side LP” in the optimization theory. Their solution paths can be generated with certain simplex algorithms. This work exploits the connection between the family of regularization methods and the parametric LP theory and lays out a general simplex algorithm and its variant for generating regularized solution paths for the feature selection problems. The significance of such algorithms is that they allow a complete exploration of the model space along the paths and provide a broad view of persistent features in the data. The implications of the general path-finding algorithms are outlined for various statistical procedures, and they are illustrated with numerical examples.

Committee: Yoonkyung Lee (Advisor); Prem Goel (Committee Member); Tao Shi (Committee Member) Subjects: Statistics