Ordinal Hyperplane Loss, Bob Vanderheyden

Doctor of Data Science and Analytics Dissertations

This research presents the development of a new framework for analyzing ordered class data, commonly called “ordinal class” data. The focus of the work is the development of classifiers (predictive models) that predict classes from available data. Ratings scales, medical classification scales, socio-economic scales, meaningful groupings of continuous data, facial emotional intensity and facial age estimation are examples of ordinal data for which data scientists may be asked to develop predictive classifiers. It is possible to treat ordinal classification like any other classification problem that has more than two classes. Specifying a model with this strategy does not fully utilize …

Information Extraction From Biomedical Text Using Machine Learning, Deepti Garg

Master's Projects

Inadequate drug experimental data and the use of unlicensed drugs may cause adverse drug reactions, especially in pediatric populations. Every year the U.S. Food and Drug Administration approves human prescription drugs for marketing. The labels associated with these drugs include information about clinical trials and drug response in pediatric population. In order for doctors to make an informed decision about the safety and effectiveness of these drugs for children, there is a need to analyze complex and often unstructured drug labels. In this work, first, an exploratory analysis of drug labels using a Natural Language Processing pipeline is performed. Second, …

Reservoir Computing In An Evolutionary Neuromorphic Framework, John J. Reynolds

Doctoral Dissertations

Neuromorphic computing is an emerging hardware paradigm for doing non-traditional computing. It has advantages over typical von Neumann systems in a myriad of different situations. In particular, it offers attractive power savings over traditional hardware, by doing spiking neural network computations. However, programming a neuromorphic spiking system is very challenging, and thus an active field of research. This work explores using the TENNLab group's neuromorphic computing framework with reservoir computing, a method for utilizing either spiking or non-spiking neural networks as dynamical systems (called reservoirs) to filter and map information from one dimension to another to form useful intermediate data …

Wheelchair Propulsion For Everyday Manual Wheelchair Users: Repetition Training And Machine Learning-Based Monitoring, Pin-Wei Chen

Arts & Sciences Electronic Theses and Dissertations

Upper limb pain and injuries are prevalent among manual wheelchair users and can restrict their participation and daily activities. Due to the high repetition and force in wheelchair propulsion, chronic wheelchair propulsion has been linked to the risk of upper limb pain and injury. Prevention of upper limb pain and injury is a high priority in wheelchair-related research. Decades of research in wheelchair propulsion biomechanics have led to clinical practice guidelines (CPG). Unfortunately, a decade after the publication of the CPG, CPG-recommended propulsion is still uncommon. Hence, for the first aim, a randomized controlled trial pilot study with two groups …

Finding A Viable Neural Network Architecture For Use With Upper Limb Prosthetics, Maxwell Lavin

Master of Science in Computer Science Theses

This paper attempts to answer the question of if it’s possible to produce a simple, quick, and accurate neural network for the use in upper-limb prosthetics. Through the implementation of convolutional and artificial neural networks and feature extraction on electromyographic data different possible architectures are examined with regards to processing time, complexity, and accuracy. It is found that the most accurate architecture is a multi-entry categorical cross entropy convolutional neural network with 100% accuracy. The issue is that it is also the slowest method requiring 9 minutes to run. The next best method found was a single-entry binary cross entropy …

Robot Motion Planning In Dynamic Environments, Hao-Tien Lewis Chiang

Computer Science ETDs

Robot motion planning in dynamic environments is critical for many robotic applications, such as self-driving cars, UAVs and service robots operating in changing environments. However, motion planning in dynamic environments is very challenging as this problem has been shown to be NP-Hard and in PSPACE, even in the simplest case. As a result, the lack of safe, efficient planning solutions for real-world robots is one of the biggest obstacles for ubiquitous adoption of robots in everyday life. Specifically, there are four main challenges facing motion planning in dynamic environments: obstacle motion uncertainty, obstacle interaction, complex robot dynamics and noise, and …

Comparative Analysis Based On Survey Of Ddos Attacks’ Detection Techniques At Transport, Network, And Application Layers, Mustafa Khambatta

Culminating Projects in Information Assurance

Distributed Denial of Service (DDOS) is one of the most prevalent attacks and can be executed in diverse ways using various tools and codes. This makes it very difficult for the security researchers and engineers to come up with a rigorous and efficient security methodology. Even with thorough research, analysis, real time implementation, and application of the best mechanisms in test environments, there are various ways to exploit the smallest vulnerability within the system that gets overlooked while designing the defense mechanism. This paper presents a comprehensive survey of various methodologies implemented by researchers and engineers to detect DDOS attacks …

An Application Of Clustering And Cluster Update Methods To Boiler Sensor Prediction And Case-Based-Reasoning To Boiler Repair, Timothy Edward Rooney

Theses and Dissertations

Driven by demand from both consumers and manufacturers alike, Internet of Things (IoT)

capabilities are being built into more products. Consumers want more control and access to their

devices, while manufacturers can find data gathered from IoT-capable products invaluable. In

this thesis, we use data from a growing fleet of IoT-connected boilers in the residential, lightcommercial, and medium-commercial ranges to demonstrate a framework for cluster initialization

and updating. We compare two methods of dynamically updating clusters: a sequential method

inspired by sequential K-means clustering and a cohesion-based method called DYNC. A predictive

artificial neural network system demonstrates the effectiveness of …

Recommending Privacy Settings For Internet-Of-Things, Yangyang He

All Dissertations

Privacy concerns have been identified as an important barrier to the growth of IoT. These concerns are exacerbated by the complexity of manually setting privacy preferences for numerous different IoT devices. Hence, there is a demand to solve the following, urgent research question: How can we help users simplify the task of managing privacy settings for IoT devices in a user-friendly manner so that they can make good privacy decisions?

To solve this problem in the IoT domain, a more fundamental understanding of the logic behind IoT users’ privacy decisions in different IoT contexts is needed. We, therefore, conducted a …

Change In Organizations, What Do They Have To Say About It? Machine Learning Testing Of An Affective Behavioral Circumplex Model Of Reactions To Change, Tiffany N. Cooper

All Dissertations

The ability for organizations to effectively systematically change their culture is becoming increasingly necessary. These changes are often implemented through a strategic process to which employee reactions have a great impact on their success. This study tested a new affective behavioral circumplex model of reactions to change. Although that was not fully supported, the data clusters that did emerge held true across samples. Not only did this study test this new model but also used new methods in Machine learning to examine qualitative responses which were found to be accurate and reliable. Furthermore, this study examined how this model is …

Visual Speech Recognition Using A 3d Convolutional Neural Network, Matthew Rochford

Master's Theses

Main stream automatic speech recognition (ASR) makes use of audio data to identify spoken words, however visual speech recognition (VSR) has recently been of increased interest to researchers. VSR is used when audio data is corrupted or missing entirely and also to further enhance the accuracy of audio-based ASR systems. In this research, we present both a framework for building 3D feature cubes of lip data from videos and a 3D convolutional neural network (CNN) architecture for performing classification on a dataset of 100 spoken words, recorded in an uncontrolled envi- ronment. Our 3D-CNN architecture achieves a testing accuracy of …

Vehicle Distance Detection Using Monocular Vision And Machine Learning, Maryam Samir Naguib Girgis Hanna

Electronic Theses and Dissertations

With the development of new cutting-edge technology, autonomous vehicles (AVs) have become the main topic in the majority of the automotive industries. For an AV to be safely used on the public roads it needs to be able to perceive its surrounding environment and calculate decisions within real-time. A perfect AV still does not exist for the majority of public use, but advanced driver assistance systems (ADAS) have been already integrated into everyday vehicles. It is predicted that these systems will evolve to work together to become a fully AV of the future. This thesis’ main focus is the combination …

Fractional Random Weighted Bootstrapping For Classification On Imbalanced Data With Ensemble Decision Tree Methods, Sean Charles Carter

USF Tampa Graduate Theses and Dissertations

Ensemble methods are commonly used for building predictive models for classification. Models that are unstable to perturbations in the training set, such as the decision tree, often see considerable reductions in error when grouped, using bootstrapped resamples of the training data to train many models. The non-parametric bootstrap, however, has limited efficacy when used on severely imbalanced data, especially when the number of observations of one or more classes is exceptionally small. We explore the fractional random weighted bootstrap, which randomly assigns fractional weights to observations, as an alternative resampling pro cedure in training machine learning ensembles, particularly decision tree …

Comparison Of Modern Controls And Reinforcement Learning For Robust Control Of Autonomously Backing Up Tractor-Trailers To Loading Docks, Journey Mcdowell

Master's Theses

Two controller performances are assessed for generalization in the path following task of autonomously backing up a tractor-trailer. Starting from random locations and orientations, paths are generated to loading docks with arbitrary pose using Dubins Curves. The combination vehicles can be varied in wheelbase, hitch length, weight distributions, and tire cornering stiffness. The closed form calculation of the gains for the Linear Quadratic Regulator (LQR) rely heavily on having an accurate model of the plant. However, real-world applications cannot expect to have an updated model for each new trailer. Finding alternative robust controllers when the trailer model is changed was …

Effect Of Neighborhood Approximation On Downstream Analytics, Saranya Soundar Rajan

Master's Theses

Nearest neighbor search algorithms have been successful in finding practically useful solutions to computationally difficult problems. In the nearest neighbor search problem, the brute force approach is often more efficient than other algorithms for high-dimensional spaces. A special case exists for objects represented as sparse vectors, where algorithms take advantage of the fact that an object has a zero value for most features. In general, since exact nearest neighbor search methods suffer from the “curse of dimensionality,” many practitioners use approximate nearest neighbor search algorithms when faced with high dimensionality or large datasets. To a reasonable degree, it is known …

Accelerating Graph Processing On Large-Scale Multicores, Masab Ahmad

Doctoral Dissertations

With the ever-increasing amount of data and input variations, portable performance is becoming harder to exploit on today’s architectures. Computational setups utilize single-chip processors, such as GPUs or large-scale multicores for graph analytics. Some algorithm-input combinations perform more efficiently when utilizing a GPU’s higher concurrency and bandwidth, while others perform better with a multicore’s stronger data caching capabilities. Architectural choices also occur within selected accelerators, where variables such as threading and thread placement need to be decided for optimal performance. This paper proposes a performance predictor paradigm for a heterogeneous parallel architecture where multiple disparate accelerators are integrated in an …

Automatic Inference Of Causal Reasoning Chains From Student Essays, Simon Mark Hughes

College of Computing and Digital Media Dissertations

While there has been an increasing focus on higher-level thinking skills arising from the Common Core Standards, many high-school and middle-school students struggle to combine and integrate information from multiple sources when writing essays. Writing is an important learning skill, and there is increasing evidence that writing about a topic develops a deeper understanding in the student. However, grading essays is time consuming for teachers, resulting in an increasing focus on shallower forms of assessment that are easier to automate, such as multiple-choice tests. Existing essay grading software has attempted to ease this burden but relies on shallow lexico-syntactic features …

Machine Learning Classification Of Interplanetary Coronal Mass Ejections Using Satellite Accelerometers, Kelsey Doerksen

Electronic Thesis and Dissertation Repository

Space weather phenomena is a complex area of research as there are many different variables and signatures that are used to identify the occurrence of solar storms and Interplanetary Coronal Mass Ejections (ICMEs), with inconsistencies between databases and solar storm catalogues. The identification of space weather events is important from a satellite operation point of view, as strong geomagnetic storms can cause orbit perturbations to satellites in low-earth orbit. The Disturbance storm time (Dst) and the Planetary K-index (Kp) are common indices used to identify the occurrence of geomagnetic storms caused by ICMEs, among several other signatures that are not …

Essays On Applied Machine Learning For Implied Volatility Interpolation And Artificial Counterfactuals, Pablo A. Crespo

Dissertations, Theses, and Capstone Projects

This dissertation consists of two chapters.

Chapter 1: Volatility estimates under the risk neutral density have become a much revisited topic of interest in recent years. The density proves itself a powerful tool for sentiment analysis, since its moments provide insights about expectations in price trends. A standard procedure for its extraction utilizes artificial volatility predictions to form a dense enough grid for approximating a complete probability distribution. This paper proposes two common machine learning technique variations to produce implied volatility predictions when data is very scarce. First, a model using regularization through a variation of a generalized LASSO path …

Machine Learning For Performance Aware Virtual Network Function Placement, Dimitrios Michael Manias

Electronic Thesis and Dissertation Repository

With the growing demand for data connectivity, network service providers are faced with the task of reducing their capital and operational expenses while simultaneously improving network performance and addressing the increased connectivity demand. Although Network Function Virtualization has been identified as a potential solution, several challenges must be addressed to ensure its feasibility. The work presented in this thesis addresses the Virtual Network Function (VNF) placement problem through the development of a machine learning-based Delay-Aware Tree (DAT) which learns from the previous placement of VNF instances forming a Service Function Chain. The DAT is able to predict VNF instance placements …

Whetstone Trained Spiking Deep Neural Networks To Spiking Neural Networks, Jiajia Zhao

Masters Theses

A deep neural network is a non-spiking artificial neural network which uses multiple structured layers to extract features from the input. Spiking neural networks are another type of artificial neural network which closely mimic biology with time dependent pulses to transmit information. Whetstone is a training algorithm for spiking deep neural networks. It modifies the back propagation algorithm, typically used in deep learning, to train a spiking deep neural network, by converting the activation function found in deep neural networks into a threshold used by a spiking neural network. This work converts a spiking deep neural network trained from Whetstone …

Enterprise Data Mining & Machine Learning Framework On Cloud Computing For Investment Platforms, Narasimharao V. Casturi

Computer Science Dissertations

Machine Learning and Data Mining are two key components in decision making systems which can provide valuable in-sights quickly into huge data set. Turning raw data into meaningful information and converting it into actionable tasks makes organizations profitable and sustain immense competition. In the past decade we saw an increase in Data Mining algorithms and tools for financial market analysis, consumer products, manufacturing, insurance industry, social networks, scientific discoveries and warehousing. With vast amount of data available for analysis, the traditional tools and techniques are outdated for data analysis and decision support. Organizations are investing considerable amount of resources in …

Predicting Absenteeism Of Female Students In Alabama, Funmilola Okelana

Dissertations and Theses

Abstract

Students are chronically absent when they miss at least 15 days of the school year. Past researchers have identified income and environment as factors that affect school absenteeism. Alabama is a poor state with a high crime rate. The hypothesis for this research is that the absenteeism of female students in Alabama is high. Do we reject or fail to reject this hypothesis. If we fail to reject this hypothesis, then what other factors can affect absenteeism in schools? How can we best predict the absenteeism of female students in Alabama? What is the effect of bad data on …

Multi-Lidar Placement, Calibration, And Co-Registration For Off-Road Autonomous Vehicle Operation, William Meadows

Theses and Dissertations

For autonomous vehicles, 3D, rotating LiDAR sensors are critically important towards the vehicle's ability to sense its environment. Generally, these sensors scan their environment, using multiple laser beams to gather information about the range and the intensity of the reflection from an object. For multi--LiDAR systems, the placement of the sensors determines the density of the combined point cloud. I perform preliminary research on the optimal LiDAR placement strategy for an off--road, autonomous vehicle known as the Halo project. I use simulation to generate large amounts of labeled LiDAR data that can be used to train and evaluate a neural …

Prediction Of Hierarchical Classification Of Transposable Elements Using Machine Learning Techniques, Manisha Panta

University of New Orleans Theses and Dissertations

Transposable Elements (TEs) or jumping genes are the DNA sequences that have an intrinsic capability to move within a host genome from one genomic location to another. Studies show that the presence of a TE within or adjacent to a functional gene may alter its expression. TEs can also cause an increase in the rate of mutation and can even promote gross genetic arrangements. Thus, the proper classification of the identified jumping genes is important to understand their genetic and evolutionary effects. While computational methods have been developed that perform either binary classification or multi-label classification of TEs, few studies …

Feature Selection And Analysis For Standard Machine Learning Classification Of Audio Beehive Samples, Chelsi Gupta

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

The beekeepers need to inspect their hives regularly in order to protect them from various stressors. Manual inspection of hives require a lot of time and effort. Hence, many researchers have started using electronic beehive monitoring (EBM) systems to collect critical information from beehives, so as to alert the beekeepers of possible threats to the hive. EBM collects information by applying multiple sensors into the hive. The sensors collect information in the form of video, audio or temperature data from the hives.

This thesis involves the automatic classification of audio samples from a beehive into bee buzzing, cricket chirping and …

Machine Learning Techniques As Applied To Discrete And Combinatorial Structures, Samuel David Schwartz

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Machine Learning Techniques have been used on a wide array of input types: images, sound waves, text, and so forth. In articulating these input types to the almighty machine, there have been all sorts of amazing problems that have been solved for many practical purposes.

Nevertheless, there are some input types which don’t lend themselves nicely to the standard set of machine learning tools we have. Moreover, there are some provably difficult problems which are abysmally hard to solve within a reasonable time frame.

This thesis addresses several of these difficult problems. It frames these problems such that we can …

Dynamic Prediction Of Treatment Outcomes For Recurrent Tuberculosis Patients, Nicole Hayes

Industrial Engineering Undergraduate Honors Theses

Tuberculosis (TB) is a disease that affects people around the world, especially people in underdeveloped countries. TB is one of the top ten causes of death globally so improvement in understanding diagnosis and treatment of TB affected patients could lead to major improvements in world health. This thesis research evaluated relapse patients specifically, deeming a relapse patient as one who has either been cured or completed their last treatment and then is diagnosed with TB again.

This research uses dynamic predictive modeling, based upon the random forest algorithm, to predict treatment outcomes for recurrent TB patients using demographic and follow-up …

A Machine Learning Framework For Energy Consumption Prediction, Chakara Rajan Madhusudanan

All Theses

Energy needs to be used very efficiently in today's world. With fast paced improvements in the industrial sector, demand is increasing, and energy efficiency programs become vital to reduce the energy wastage while also meeting the demand. The analysis of several scenarios used by policy makers suggest that for the global temperature to raise by less than 2° C by the end of this century, it is necessary to reduce industrial energy consumption increase by at least a half. To be on track with these scenarios and to achieve the desirable targets, it is important that we incorporate a dependable …

Static Malware Detection Using Deep Neural Networks On Portable Executables, Piyush Aniruddha Puranik

UNLV Theses, Dissertations, Professional Papers, and Capstones

There are two main components of malware analysis. One is static malware analysis and the other is dynamic malware analysis. Static malware analysis involves examining the basic structure of the malware executable without executing it, while dynamic malware analysis relies on examining malware behavior after executing it in a controlled environment. Static malware analysis is typically done by modern anti-malware software by using signature-based analysis or heuristic-based analysis.

This thesis proposes the use of deep neural networks to learn features from a malware’s portable executable (PE) to minimize the occurrences of false positives when recognizing new malware. We use the …