Digital Commons Network^™

Advanced Caching And Streaming For Large Scale Point Cloud Data Visualization On The Web, Pravin Poudel

All Graduate Theses and Dissertations, Fall 2023 to Present

Point clouds are widely used in various applications such as 3D modeling, geospatial analysis, robotics, and more. One of the key advantages of 3D point cloud data is that, unlike other data formats like texture, it is independent of viewing angle, surface type, and parameterization. Since each point in the point cloud is independent of the other, it makes it the most suitable source of data for tasks like object recognition, scene segmentation, and reconstruction. Point clouds are complex and verbose due to the numerous attributes they contain, many of which may not be always necessary for rendering, making retrieving …

Deep Learning With Effective Hierarchical Attention Mechanisms In Perception Of Autonomous Vehicles, Qiuxiao Chen

All Graduate Theses and Dissertations, Fall 2023 to Present

Autonomous vehicles need to gather and understand information from their surroundings to drive safely. Just like how we look around and understand what's happening on the road, these vehicles need to see and make sense of dynamic objects like other cars, pedestrians, and cyclists, and static objects like crosswalks, road barriers, and stop lines.

In this dissertation, we aim to figure out better ways for computers to understand their surroundings in the 3D object detection task and map segmentation task. The 3D object detection task automatically spots objects in 3D (like cars or cyclists) and the map segmentation task automatically …

Optimal Stopping Of Multi-Robot Exploration For Unknown, Bounded Environments, Trey D. Crowther

All Graduate Theses and Dissertations, Fall 2023 to Present

Limited resources and uncertainty pose a substantial problem for multi-robot exploration of unknown environments. This research paper looks to determine the optimal time to terminate robot exploration while maximizing information gathered. Whilst making this determination, the system's resources and capabilities must be taken into account. To see if our strategy works, we ran many simulations in varying environments. The results of this research are important for real-world uses like robot exploration, search and rescue missions, and automated surveillance. Determining when to stop exploring can help the system save resources, explore faster, and make better decisions.

Analysis Of Student Behavior And Score Prediction In Assistments Online Learning, Aswani Yaramala

All Graduate Theses and Dissertations, Fall 2023 to Present

Understanding and analyzing student behavior is paramount in enhancing online learning, and this thesis delves into the subject by presenting an in-depth analysis of student behavior and score prediction in the ASSISTments online learning platform. We used data from the EDM Cup 2023 Kaggle Competition to answer four key questions. First, we explored how students seeking hints and explanations affect their performance in assignments, shedding light on the role of guidance in learning. Second, we looked at the connection between students mastering specific skills and their performance in related assignments, giving insights into the effectiveness of curriculum alignment. Third, we …

Solar Flare Prediction From Extremely Imbalanced Multivariate Time Series Data Using Minimally Random Convolutional Kernel Transform, Kartik Saini

All Graduate Theses and Dissertations, Fall 2023 to Present

Solar flares are characterized by sudden bursts of electromagnetic radiation from the Sun's surface, and caused by the changes in magnetic field states in solar active regions. Earth and its surrounding space environment can suffer from various negative impacts caused by solar flares ranging from electronic communication disruption to radiation exposure-based health risks to the astronauts. In this paper, we address the solar flare prediction problem from magnetic field parameter-based multivariate time series (MVTS) data using multiple state-of-the-art machine learning classifiers that include MINImally RandOm Convolutional KErnel Transform (MINIROCKET), Support Vector Machine (SVM), Canonical Interval Forest (CIF), Multiple Representations SEQuence …

Collaborative Task Completion For Simulated Hexapod Robots Using Reinforcement Learning, Tayler Don Baker

All Graduate Theses and Dissertations, Fall 2023 to Present

There is growing interest in developing autonomous systems capable of exhibiting collaborative behaviors. Using methods such as reinforcement learning is another way to train multiple robots for collaborative task completion. This study was able to successfully in simulation train multiple hexapod robots to push a target to a designated goal collaboratively. This required each robot to learn how find the target and push that target to a goal. This work suggests that using reinforcement learning for collaborative task completion for hexapod robots may simplify the complexity of the software and improve the decisions that they make.

Adversarially Reweighted Sequence Anomaly Detection With Limited Log Data, Kevin Vulcano

All Graduate Theses and Dissertations, Fall 2023 to Present

In the realm of safeguarding digital systems, the ability to detect anomalies in log sequences is paramount, with applications spanning cybersecurity, network surveillance, and financial transaction monitoring. This thesis presents AdvSVDD, a sophisticated deep learning model designed for sequence anomaly detection. Built upon the foundation of Deep Support Vector Data Description (Deep SVDD), AdvSVDD stands out by incorporating Adversarial Reweighted Learning (ARL) to enhance its performance, particularly when confronted with limited training data. By leveraging the Deep SVDD technique to map normal log sequences into a hypersphere and harnessing the amplification effects of Adversarial Reweighted Learning, AdvSVDD demonstrates remarkable efficacy …

On The Computability Of Primitive Recursive Functions By Feedforward Artificial Neural Networks, Vladimir A. Kulyukin

Computer Science Faculty and Staff Publications

We show that, for a primitive recursive function h(x, t), where x is a n-tuple of natural numbers and t is a natural number, there exists a feedforward artificial neural network 𝔑(x, t), such that for any n-tuple of natural numbers z and a positive natural number m, the first m + 1 terms of the sequence {h(z, t)} are the same as the terms of the tuple (𝔑(z, 0), ... ,𝔑(z, m)).

Contemporary Art Authentication With Large-Scale Classification, Todd Dobbs, Abdullah-Al-Raihan Nayeem, Isaac Cho, Zbigniew Ras

Computer Science Faculty and Staff Publications

Art authentication is the process of identifying the artist who created a piece of artwork and is manifested through events of provenance, such as art gallery exhibitions and financial transactions. Art authentication has visual influence via the uniqueness of the artist’s style in contrast to the style of another artist. The significance of this contrast is proportional to the number of artists involved and the degree of uniqueness of an artist’s collection. This visual uniqueness of style can be captured in a mathematical model produced by a machine learning (ML) algorithm on painting images. Art authentication is not always possible …

A Novel Fuzzy Relative-Position-Coding Transformer For Breast Cancer Diagnosis Using Ultrasonography, Yanhui Guo, Ruquan Jiang, Xin Gu, Heng-Da Cheng, Harish Garg

Computer Science Faculty and Staff Publications

Breast cancer is a leading cause of death in women worldwide, and early detection is crucial for successful treatment. Computer-aided diagnosis (CAD) systems have been developed to assist doctors in identifying breast cancer on ultrasound images. In this paper, we propose a novel fuzzy relative-position-coding (FRPC) Transformer to classify breast ultrasound (BUS) images for breast cancer diagnosis. The proposed FRPC Transformer utilizes the self-attention mechanism of Transformer networks combined with fuzzy relative-position-coding to capture global and local features of the BUS images. The performance of the proposed method is evaluated on one benchmark dataset and compared with those obtained by …

A Neural-Network-Based Landscape Search Engine: Lse Wisconsin, Matthew Haffner, Matthew Dewitte, Papia F. Rozario, Gustavo A. Ovando-Montejo

Environment and Society Faculty Publications

The task of image retrieval is common in the world of data science and deep learning, but it has received less attention in the field of remote sensing. The authors seek to fill this gap in research through the presentation of a web-based landscape search engine for the US state of Wisconsin. The application allows users to select a location on the map and to find similar locations based on terrain and vegetation characteristics. It utilizes three neural network models—VGG16, ResNet-50, and NasNet—on digital elevation model data, and uses the NDVI mean and standard deviation for comparing vegetation data. The …

Comparative Study Of Clustering Techniques On Eye-Tracking In Dynamic 3d Virtual Environments, Scott Johnson

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Eye-tracking has been used for decades to understand how and why an individual focuses on particular objects, areas, and elements of space. A vast body of knowledge exists on how eye-tracking is measured. However, historically, eye-tracking has been predominately studied using 2D environments, with limited work in 3D environments. The purpose of this study is to identify which methods most accurately represent the areas that have captured the participant’s visual attention within a 3D dynamic environment. This will be completed by evaluating different clustering methods of fixations using a customized virtual reality tool that collects eye-tracking data. There exist several …

Physics-Guided Deep Learning For Solar Wind Modeling At L1 Point, Robert M. Johnson

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Neural networks are adept at finding patterns that are too long and too small for humans to find in data. Usually, this power is used to generate predictions with greater accuracy than most alternative models. However, we can also use this power to understand more about the data we train these networks on. We do this by changing the data that the networks train on and the data they are tested on. This allows us to both control the maximum length of a pattern and to compare data between different groups, in our case, different solar cycles. This thesis is …

Generalizing Deep Learning Methods For Particle Tracing Using Transfer Learning, Shubham Gupta

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Particle tracing is a very important method for scientific visualization of vector fields, but it is computationally expensive. Deep learning can be used to speed up particle tracing, but existing deep learning models are domain-specific. In this work, we present a methodology to generalize the use of deep learning for particle tracing using transfer learning. We demonstrate the performance of our approach through a series of experimental studies that address the most common simulation design scenarios: varying time span, Reynolds number, and problem geometry. The results show that our methodology can be effectively used to generalize and accelerate the training …

Constrained Route Optimization With Fleet Considerations For Electrified Heavy-Duty Freight Vehicles, Zarin Subah Shamma

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Almost 75% of traffic-related emissions are caused by heavy-duty freight trucks and significantly impact neighborhoods, schools, and communities around shipping and distribution lines. With poor air quality and respiratory health, many children in at-risk and disadvantaged communities experience high rates of asthma, lower attendance in school, and lower concentration. This research creates to improve the impacts of heavy-duty electric freight by improving the route efficiency (in terms of energy, time, or route distance) of EV trucks. Our software and algorithms are tested in a simulation environment using data from several thousand fleet trucks operating in the Salt Lake City area. …

Proxy Voting Coordination Mechanisms: Determining How Agents Should Coordinate In A Continuous Preference Space, Michael D. Hegerhorst

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Illness, injury, and other impediments are common occurrences of everyday life. Such impediments prevent or deter voters from participating in important parts of the voting process, especially deliberation, bargaining, and the voting itself. Without participation, the results of the vote may change. There is a need to provide a system in which voters are still able to participate in important voting processes to ensure their vote is represented. We explore ‘proxy voting,’ a system in which voters are able to select another individual, or proxy, to vote on their behalf. By choosing a good proxy, a voter can still …

Accuracy Vs. Energy: An Assessment Of Bee Object Inference In Videos From On-Hive Video Loggers With Yolov3, Yolov4-Tiny, And Yolov7-Tiny, Vladimir A. Kulyukin, Aleksey V. Kulyukin

Computer Science Faculty and Staff Publications

A continuing trend in precision apiculture is to use computer vision methods to quantify characteristics of bee traffic in managed colonies at the hive's entrance. Since traffic at the hive's entrance is a contributing factor to the hive's productivity and health, we assessed the potential of three open-source convolutional network models, YOLOv3, YOLOv4-tiny, and YOLOv7-tiny, to quantify omnidirectional traffic in videos from on-hive video loggers on regular, unmodified one- and two-super Langstroth hives and compared their accuracies, energy efficacies, and operational energy footprints. We trained and tested the models with a 70/30 split on a dataset of 23,173 flying bees …

On Correspondences Between Feedforward Artificial Neural Networks On Finite Memory Automata And Classes Of Primitive Recursive Functions, Vladimir A. Kulyukin

Computer Science Faculty and Staff Publications

When realized on computational devices with finite quantities of memory, feedforward artificial neural networks and the functions they compute cease being abstract mathematical objects and turn into executable programs generating concrete computations. To differentiate between feedforward artificial neural networks and their functions as abstract mathematical objects and the realizations of these networks and functions on finite memory devices, we introduce the categories of general and actual computabilities and show that there exist correspondences, i.e., bijections, between functions computable by trained feedforward artificial neural networks on finite memory automata and classes of primitive recursive functions.

Deep Learning With Attention Mechanisms In Breast Ultrasound Image Segmentation And Classification, Meng Xu

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Breast cancer is a great threat to women’s health. Breast ultrasound (BUS) imaging is commonly used in the early detection of breast cancer as a portable, valuable, and widely available diagnosis tool. Automated BUS image analysis can assist radiologists in making accurate and fast decisions. Generally, automated BUS image analysis includes BUS image segmentation and classification. BUS image segmentation automatically extracts tumor regions from a BUS image. BUS image classification automatically classifies breast tumors into benign or malignant categories. Multi-task learning accomplishes segmentation and classification simultaneously, which makes it more appealing and practical than an either individual task. Deep neural …

Coding Bootcamps - Perceptions And Outcomes, Logan L. Hendricks

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

This thesis is focused on gathering, aggregating and analysing data related to software development coding bootcamps. It comprises of three major research initiatives: A coding bootcamp outcomes meta-analysis, a study on perspectives regarding white-label coding bootcamps, and the data analysis of a survey gathering long-term outcomes of coding bootcamp and certificate program graduates.

The first study aggregates graduate outcome data from the three main organizations that review coding bootcamp outcomes: CourseReport.com, SwitchUp.com and the Council on Integrity in Results Reporting (CIRR). The purpose of this meta-review is to establish a baseline dataset which is immediately utilized in my further research. …

Adversarial Swarming: A Groundwork For Multi-Drone Independent Interception Exercises Through Ma-Poca In Unity, Johnathan D. Kunz

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

As drones become more popular and easier to use, air spaces are becoming more congested. Airports, hospitals, and similar structures require controlled, safe airspaces and drones are increasingly a threat. Locally controlled airspace requires efficient removal of airborne threats to continue sensitive operations. Many methods have been investigated for removing drones from contested airspace. Generally these methods involve ground-based signal disruption, physical contact, or drone interception of a single intruder. In this work we present a drone interception model with a low-cost, low-capability group of short-range drones intercepting an incoming drone.

A Generative Neural Network For Discovering Near Optimaldynamic Inductive Power Transfer Systems, Md Shain Shahid Chowdhury Oni

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

An urgent need is to electrify transportation to lower carbon emissions into the atmosphere. Wireless charging makes electrical vehicles (EVs) more convenient and cheaper because energy is transferred to the vehicle without the need to plug it in. Dynamic wireless charging is particularly interesting, where the vehicle does not need to stop to receive the energy. This technology requires the EV and the roadway to include coils of wire, where the roadway coil is energized as the vehicle passes over it to induce an electrical current in the EV coil through electromagnetic induction. However, the problem of designing the two …

Algorithms For Unit-Disk Graphs And Related Problems, Yiming Zhao

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

In this dissertation, we study algorithms for several problems on unit-disk graphs and related problems. The unit-disk graph can be viewed as an intersection graph of a set of congruent disks. Unit-disk graphs have been extensively studied due to many of their applications, e.g., modeling the topology of wireless sensor networks. Lots of problems on unit-disk graphs have been considered in the literature, such as shortest paths, clique, independent set, distance oracle, diameter, etc. Specifically, we study the following problems in this dissertation: L₁ shortest paths in unit-disk graphs, reverse shortest paths in unit-disk graphs, minimum bottleneck moving spanning …

Generative Neural Network Approach To Designing And Optimizing Dynamic Inductive Power Transfer Systems, Andrew Pond Curtis

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Electric vehicles (EVs) offer many improvements over traditional combustion engines including increasing efficiency, while decreasing cost of operation and emissions. There is a need for the development of cheap and efficient charging systems for the future success of EVs. Most EVs currently utilize static plug-in charging systems. An alternative charging method of significant interest is dynamic inductive power transfer systems (DIPT). These systems utilize two coils, one placed in the vehicle and one in the roadway to wirelessly charge the vehicle as it passes over. This method removes the current limitations on EVs where they must stop and statically charge …

Rethinking Integrated Computer Science Instruction: A Cross-Context And Expansive Approach In Elementary Classrooms, Umar Shehzad, Jody E. Clarke-Midura, Kimberly Beck, Jessica F. Shumway, Mimi M. Recker

Publications

This study examines how a rural-serving school district aimed to provide elementary level computer science (CS) by offering instruction during students’ computer lab, a class taught by paraprofessional educators with limited background in computing. As part of a research practice partnership, cross-context mathematics and CS lessons were co-designed to expansively frame and highlight connections across – as opposed to integration within – the two subjects. Findings indicate that the paraprofessionals teaching the lessons generally reported positive experiences and understanding of content; however, those less comfortable with the content reported lower student interest. Further, most students who engaged with the lessons …

Geometry And Coding: Introducing An Interactive And Integrated Mathematics-Computer Science Unit, Kimberly Beck, Jessica F. Shumway

Publications

As part of a collaborative project between Utah State University, the Cache County School District, and Stanford, instructional units were designed for fifth-grade students. These units integrated math concepts of geometrical shapes and computer science concepts of sequences, conditionals, and loops. One component of the unit was implemented in math classrooms by math teachers, and the other component was implemented in computer labs. This presentation will focus on the math unit as presented at the National Council of Teachers of Mathematics (NCTM-V).

Plagiarism Deterrence In Cs1 Through Keystroke Data, Kaden Hart, Chad Mano, John M. Edwards

Computer Science Student Research

Recent work in computing education has explored the idea of analyzing and grading using the process of writing a computer program rather than just the final submitted code. We build on this idea by investigating the effect on plagiarism when the process of coding, in the form of keystroke logs, is submitted for grading in addition to the final code. We report results from two terms of a university CS1 course in which students submitted keystroke logs. We find that when students are required to submit a log of keystrokes together with their written code they are less likely to …

Accurate Estimation Of Time-On-Task While Programming, Kaden Hart, Christopher M. Warren, John Edwards

Computer Science Student Research

In a recent study, students were periodically prompted to self-report engagement while working on computer programming assignments in a CS1 course. A regression model predicting time-on-task was proposed. While it was a significant improvement over ad-hoc estimation techniques, the study nevertheless suffered from lack of error analysis, lack of comparison with existing methods, subtle complications in prompting students, and small sample size. In this paper we report results from a study with an increased number of student participants and modified prompting scheme intended to better capture natural student behavior. Furthermore, we perform a cross-validation analysis on our refined regression model …

Ambient Electromagnetic Radiation As A Predictor Of Honey Bee (Apis Mellifera) Traffic In Linear And Non-Linear Regression: Numerical Stability, Physical Time And Energy Efficiency, Vladimir Kulyukin, Daniel Coster, Anastasiia Tkachenko, Daniel Hornberger, Aleksey V. Kulyukin

Computer Science Faculty and Staff Publications

Since bee traffic is a contributing factor to hive health and electromagnetic radiation has a growing presence in the urban milieu, we investigate ambient electromagnetic radiation as a predictor of bee traffic in the hive’s vicinity in an urban environment. To that end, we built two multi-sensor stations and deployed them for four and a half months at a private apiary in Logan, Utah, U.S.A. to record ambient weather and electromagnetic radiation. We placed two non-invasive video loggers on two hives at the apiary to extract omnidirectional bee motion counts from videos. The time-aligned datasets were used to evaluate 200 …

Co-Designing Elementary-Level Computer Science And Mathematics Lessons: An Expansive Framing Approach, Umar Shehzad, Jody Clarke-Midura, Kimberly Beck, Jessica Shumway, Mimi Recker

Publications

This study examines how a rural-serving school district aimed to provide elementary-level computer science (CS) by offering instruction during students’ computer lab time. As part of a research-practice partnership, cross-context mathematics and CS lessons were co-designed to expansively frame and highlight connections across – as opposed to integration within – the two subjects. Findings indicated that most students who engaged with the lessons across the lab and classroom contexts reported finding the lessons interesting, seeing connections to their mathematics classes, and understanding the programming. In contrast, a three-level logistic regression model showed that students who only learned about mathematics connections …