Physical Sciences and Mathematics Commons^™

Ur-60 Video-To-Video Synthesis With Semantically Segmented Video, Aydan Mufti, Jordan S Hasty

C-Day Computing Showcase

Our project involves studying the usage of generative adversarial networks (GANs) to translate semantically segmented video to photo-realistic video in a process known as video-to-video synthesis. The model is able to learn a mapping from semantically segmented masks to real-life images which depict the corresponding semantic labels. To achieve this, we employ a conditional GAN-based learning method that produces output conditionally based on the source video to be translated. Our model is capable of synthesizing a translated video, given semantically labeled video, that resembles real video by accurately replicating low-frequency details from the source.Advisors(s): Dr. Mohammed AledhariTopic(s): Artificial IntelligenceCS 4732

Ur-63 Low Cost High Impact Fall Detection At The Edge, Dylan Sirna, Noah Sage

C-Day Computing Showcase

ML models have become more accurate, powerful and portable in recent years, the purpose of this project is to explore how these advances can be applied towards fall detection for less cost than before possible. This project explores the application of micro controllers which have become cheaper and stronger along with emerging machine learning models that can be trained on a traditional computer with greater resources and then port the model to be interpreted on a micro-controller such as a raspberry pi. These two factors lead to the reason to revisit the problem of fall detection, a problem that plagues …

Gc-52 Ksu Spring Capstone - It Security Solution For Small Business – Group 1, William Simmons, Olajumoke Giwa, Beau Beard, Collin Peters

C-Day Computing Showcase

For this project, we decided to use the nopCommerce open-source eCommerce solution for our simulated small business. Leveraging the pre-configured web server provide to our group, we installed the nopCommerce solution package and built our security program around the web site. Our security program consists of nginx configured web server and load balancer, SSL certificate for encryption, firewall and a secure backend SQL server.Advisors(s): Project Sponsor: Dr. Lei Li Course Professor: Dr. Meng HanTopic(s): SecurityIT 7993

Gr-40 Design And Implementation Of A Microservices Web-Based Architecture For Code Deployment And Testing, Soin Abdoul Kassif Baba M Traore

C-Day Computing Showcase

Many tech stars like Netflix, Amazon, PayPal, eBay, and Twitter are evolving from monolithic to a microservice architecture due to the benefits for Agile and DevOps teams. Microservices architecture can be applied to multiple industries, like IoT, using containerization. Virtual containers give an ideal environment for developing and testing IoT technologies. Since the IoT industry has exponential growth, it is the responsibility of universities to teach IoT with hands-on labs to minimize the gap between what the students learn and what is on-demand in the job market. That can be done by using containerization. There are many approaches in the …

Gr-44 An Efficient Intrusion Detection Framework Based On Federated Learning For Iot Networks., Osama Shahid

C-Day Computing Showcase

There are abundant number of IoT devices that are connected on over multiple networks. These devices can be exposed to multiple different types of network threats. Though, these devices do have security and software that does act as a wall of protection we purpose a Federated Learning (FL) approach that would allow detection of threats of a network for IoT devices. Federated Learning can be best described as decentralized training. Adhering to the GDPR rules that prevent data from being distributed, FL addressed the challenge by bringing the ML model to the data rather than the traditional method where the …

Uc-26 Pose Extraction For Real-Time Workout Assist_Capstone Group_W01_Spring Semester, Royce Camp, Zach Christmas, Jonathon Segars, Amanda Mead, Cameron Page

C-Day Computing Showcase

Motion Intelligence Research Project We installed and tested many existing pose extraction technologies in many situations. We provided reports on the different software solutions and decided on a single solution that performed the best. We will extract key-points and track the movements across multiple dimensions. We will demonstrate the X and Y movements of everyone for our chosen software solution in a Business Intelligent tool (PowerBI). Given the current epidemic, we are not going to be able to compare against the professional KinaTrax software on campus. Although, Dr. Xie has given us other software packages to compare our given software …

Uc-35 Development Of An Automated Software Packaging Solution For Linux, Robert D Ryan, Sammi Figueroa, Dylan Parker, Blair Hill, Bishwo R Marhatta

C-Day Computing Showcase

The main problem with using Linux software in the science and Bioinformatics community is because Linux has a large number of distributions and dependencies. This hinders researches and science students with the problem of tracking down dependencies for software which could then further break the existing system dependencies. Our team looked to solve these problems by creating a BASH script that could quickly mass package AppImages and contain Linux software with all dependencies. Our team worked through the last ten weeks and researched all components of AppImage and discovered all means to more easily package and have a repeatable process …

Gr-67 Representation Learning For Motion Sequence, Saisangararamaleengam Alagapan, Alexandru Malos, Roshni Kishor, Venkateswara Reddy Mosali

C-Day Computing Showcase

This research project proposes a new deep learning architecture that is used to align human poses to be used in an exercise assistant system. In short, the assistant system takes a video feed of a user doing exercise, then provides visual feedback by comparing the user’s current pose to a professional trainer’s pose that is stored in the system. We design a new deep architecture to accomplish this task and show better accuracy and efficiency.Advisors(s): Project Sponsors - Dr. Ying Xie & Dr. Linh Le Project Advisor- Dr. Meng HanTopic(s): Data/Data AnalyticsIT 7993

Gr-70 Defending Data Reconstruction Through Adaptive Image Augmentation, Seunghyeon Shin

C-Day Computing Showcase

In this paper, we introduce a data augmentation-based defense strategy for preventing the reconstruction of training data through the exploitation of stolen model gradient. The collection of training data to a centralized server has been required for the training of neural networks in traditional machine learning. However, as privacy becomes a significant concern, the concept of Federated learning is introduced. In federated learning, a centralized server shares the well-trained neural network and participating end-users send the gradient back to the server after training without sharing the sensitive data itself. As the concept of federated learning does not share the original …

Uc-36 Using Machine Learning Techniques To Predict Rt-Pcr Results For Covid-19 Patients., Bradley T Durden, Mathew Shulman, Andy Reynolds, Thomas A Phillips, Demontae Moore, Indya Andrews

C-Day Computing Showcase

With the COVID-19 pandemic still a threat, healthcare professionals and medical industries keep searching for better ways to mitigate the spread of COVID-19. While Machine Learning has been applied in many other domains, there is now a high demand for diagnosis systems that utilize Machine Learning techniques in the healthcare domain and in particular combating COVID-19. In this project, we explore the role of Machine Learning models in combating COVID-19, using WEKA as the main tool for analysis.Advisors(s): Dr. Ming Yang - IT 4983 Capstone Professor Dr. Seyedamin Pouriyeh - Project OwnerTopic(s): Data/Data AnalyticsIT 4983

Uc-39 Journalistic Integrity Vis Artifical Intelligence, Dylan Dalton, Ray Martin, Duy Nquyen, Yesse Quezada, Brian Dominguez

C-Day Computing Showcase

We are developing a web app to recognize and rate political bias in online journalism using artificial intelligence. All human writing inherently contains bias ,however bias is less harmful if it is transparent to the reader because they can now make informed decisions about what they read. We've collected articles and and reactions to them from online sources, and then used Neural Networks trained for natural language processing to determine bias. The project can predict bias labels on a news articles with 82% accuracy.Advisors(s): Reza Meimandi Parizi - course instructor Asher Nuckolls - project ownerTopic(s): Artificial IntelligenceSWE4724

Gc-28 Modern Web Scraping, Kenny Randolph, Joselyn Giron, Denise Tucker, Justin B Bridges, Sandhya Bantu

C-Day Computing Showcase

This project was developed for the IT7993 Capstone class in the May semester of 2021.The goal of the project is to scrape all names of key professionals of organizations in the open990.org website and insert that information into a structured database for query and analysis. The Key Professionals dataset aims to include global coverage of key investor and consultant professionals, beginning with US-based companies, involved in making an investment decision.   The overarching aim of this project is to create a one-stop center for institutional asset management distribution intelligence; the one spot to go for mandates, documentation and profiles of consultants, …

Gr-33 Efficient Yet Robust Privacy Preservation \\For Mpeg-Dash Based Video Streaming, Luke A Cranfill

C-Day Computing Showcase

MPEG-DASH is a video streaming standard that outlines protocols for sending audio and video content from a server to a client over HTTP. However, it creates an opportunity for an adversary to invade users' privacy. While a user is watching a video, information is leaked in the form of meta-data, the size and time that the server sent data to the user. After a fingerprint of this data is created, the adversary can use this to identify whether a target user is watching the corresponding video. Only one defense strategy has been proposed to deal with this problem: differential privacy …

Gr-64 How Could Iot Assist Healthcare System During Covid-19 Or Future Pandemics?, Mohammad Nasajpour Esfahani

C-Day Computing Showcase

The Internet of Things (IoT), a technology built upon sensors and devices, has shown great applicability among various domains, especially healthcare. This pandemic has critically impacted all parts of society including people, health centers, businesses, authorities, etc. Researchers are attempting to adopt different technologies to mitigate this virus faster and save more lives. Regarding the great benefits that Internet of Things (IoT) has brought into different areas within the healthcare domain, this technology has been performing several main tasks including diagnosing, monitoring, tracing, disinfecting, and vaccinating to combat this virus. Our research is conducted of the possible IoT solutions to …

Uc-12 Comprehensive Security Solution For Small E-Commerce Business, Patrick Mccollums, Hristo Bakalov, Philinda Morse, Tyler Phillips, Watson Day

C-Day Computing Showcase

Project Description: Create an e-commerce server and a comprehensive security program to protect a web server for a simulated small business. This server will include security tools such as intrusion detection, firewall, and network monitoring. The installation and maintenance of this solution will be documented as part of the final documentation package. The server will be reviewed for exploitation from other teams while we attempt the exploitation of their server(s). Research/Motivation: How to research, install, configure, and integrate various open-source software packages for information security, e-commerce, web hosting, and database. Our motivation for this project was to create and secure …

Uc-15 Malware Analysis Using Reverse Engineering, Shamour Jones, Cynthia S Marcellus, Andy Pham, Nathan Rowe, Joshua Rowland

C-Day Computing Showcase

The motivation for this project is driven by evaluation of the different tools on the market that allow for breaking down executables or binary files, and understanding what the malware is doing. By reverse-engineering the malware, we can understand its impact and how to protect against it. Our focus is to understand where different tools are stronger than others, as well as understand the evolving landscape of malware and security overall. For this capstone project, we utilized two different tools and many sample malware files. The methods used to debug the malware are detailed in our milestone two report and …

Uc-25 Woodline Interiors Project Planning Application, Beniamin Costea, Iram Nawaz, Natan Beraki, Daniel Lopez, Carter Richter, Yeonkuk Woo

C-Day Computing Showcase

Meeting with the client regularly, we’ve established one of the many solutions to their requirements. The client, WoodLine Interiors, needs a well-designed system that can solve product management, client communication, and project management. The main issue that the company is facing is in the area of managing multiple projects and tracking project progress. Their issue is commonly addressed by a large number of companies that create software solutions, but it’s not personalized for their appropriate needs in the field of cabinetry. As a result our solution created the perfect stages and states of the project so there is no more …

Uc-31 An Empirical Study Of Thermal Attacks On Edge Platforms, Tyler Holmes, Justin Duchatellier

C-Day Computing Showcase

Cloud-edge systems are vulnerable to thermal attacks as the increased energy consumption may remain undetected, while occurring alongside normal, CPU-intensive applications. The purpose of our research is to study thermal effects on modern edge systems. We also analyze how performance is affected from the increased heat and identify preventative measures. We speculate that due to the technology being a recent innovation, research on cloud-edge devices and thermal attacks is scarce. Other research focuses on server systems rather than edge platforms. In our paper, we use a Raspberry Pi 4 and a CPU-intensive application to represent thermal attacks on cloud-edge systems. …

Uc-62 Machine Learning: Twitter Bots In Disguise, Matthew Joseph Scheer, Nicolas Vasquez, James C Andersen, Joshua Tiangco, Justin Van, Cody R Walicek, Daniel Rimmel

C-Day Computing Showcase

This project was designed to help fight against misinformation spread by bots(computers), the goal assigned to us was to find and inform Twitter users of bots that follow and are being followed by the user.Advisors(s): Dr. Reza PariziTopic(s): Artificial IntelligenceSWE 4724

The Potential Of Carbon Capture Through Mineral Weathering, Noah Planavsky

Sustainability Seminar Series

Noah Planavsky is an Associate Professor in the Department of Earth and Planetary Sciences. He joined the faculty in 2012 after doing graduate work at University of California, Riverside. He is an isotope geochemist that works on environmental change in Earth’s past, present, and future. His work combines field studies, analytical chemistry, and geochemical modeling. He has worked extensively on atmospheric evolution—with a particular focus on changes in oxygen and carbon dioxide concentrations. Current projects focus on changes in ocean oxygen levels and on the potential for carbon capture through enhanced mineral weathering in marine and terrestrial environments.

An Assessment Of The Spatial And Temporal Distribution Of Microplastics In Surface And Subsurface Sediment Of Lake Huron, North America, Sara L. Belontz

Electronic Thesis and Dissertation Repository

The awareness of and the data on the prevalence of microplastic (plastic particles <5mm) pollution in freshwater environments is rapidly increasing, as low-degrading polymers are being detected in various environmental matrices of the Laurentian Great Lakes. However, the accumulation, distribution and deposition of microplastics in offshore depositional environments of the Great Lakes, and particularly Lake Huron, is relatively unknown. In this study, benthic sediment from various Lake Huron waterbodies (main basin, Georgian Bay, the North Channel, and Saginaw Bay) was quantified for microplastic particles (fibres, fragments, films, and beads). The North Channel contained the greatest microplastic abundances, averaging 47,398 particles per kg^-1 dry weight sediment (p kg^-1 dw), followed by Georgian Bay (21,390 p kg^-1 dw), the main basin (15,910 p kg^-1 dw) and Saginaw Bay (1,592 p kg^-1 dw). The results suggest that microplastic abundances in offshore settings are positively correlated with increasing water depth (p=0.004) and are controlled by lake bottom geomorphology. Hydrodynamic processes are a prevailing force driving microplastic dispersion and deposition into the offshore, in contrast to source-based drivers closer to the shoreline. Sediment cores were examined from …

The James Webb Space Telescope And Scientific Progress, Robert Astle

Quest

Independent Research Paper

Research in progress for PHYS 2425: University Physics I

Faculty Mentor: Raji Kannampuzha, Ph.D.

The following paper represents research work done by students in University Physics 2425, the first half of a two-semester introductory course in physics. It is a calculus- based physics course, intended primarily for physics, chemistry, math, and engineering majors. Students are introduced to the concept of academic research by learning to ask research-focused questions and then use the library resources to pursue outside research to find answers. For this assignment, students are asked to investigate a physical science, biological science, or technology problem …

K-Classes Of Brill–Noether Loci And A Determinantal Formula, D. Anderson, Linda Chen, N. Tarasca

Mathematics & Statistics Faculty Works

We compute the Euler characteristic of the structure sheaf of the Brill–Noether locus of linear series with special vanishing at up to two marked points. When the Brill–Noether number ρ is zero, we recover the Castelnuovo formula for the number of special linear series on a general curve; when ρ=1⁠, we recover the formulas of Eisenbud-Harris, Pirola, and Chan–Martín–Pflueger–Teixidor for the arithmetic genus of a Brill–Noether curve of special divisors. These computations are obtained as applications of a new determinantal formula for the K-theory class of certain degeneracy loci. Our degeneracy locus formula also specializes to new determinantal expressions for …

382— Wiyn Open Cluster Study: Ubvri Photometry Of Ngc 2204, Kylie Snyder, Dante Scarazzini

GREAT Day Posters

The purpose of this project was to study the open star cluster NGC2204 using images taken at Kitt Peak National Observatory using the WIYN 0.9m telescope. These images were analyzed photometrically with the intention of determining the reddening, metallicity, age, and distance modulus of the star cluster. Each image was analyzed using software that determined the point spread function and applied that function to determine the magnitude of each star in that image. These magnitudes were taken for each filter, UBVRI, and then combined and averaged to create a single catalog. Standard stars, taken on the same night, were used …

075— A Test Of The Efficacy Of The Augmented Reality Sandbox To Improve Students' Reading Topographic Map Skills, Megan Jankowiak, Alexandra Ranaudo

GREAT Day Posters

The augmented reality (AR) sandbox is a teaching tool used to connect two-dimensional (2D) and three-dimensional (3D) map representations through modeling. Our hypothesis is: the AR sandbox is effective at teaching topographic map reading skills by comparing pre vs. post test scores. Student data was processed to analyze the efficiency of using the AR sandbox to increase student ability to read topographic maps. Data consists of student answers on a topographic map reading skills test at the beginning and the end of a Historical Geology college course. Between tests, students used the AR sandbox to complete short exercises during multiple …

201— Diversity, Equity, And Inclusion Within The Department Of Geological Sciences At Suny Geneseo, Maria Leonard

GREAT Day Posters

The field of Geosciences historically has lacked diversity. This lack of diversity is observed at all levels of higher education—for example, 90% of doctoral degrees are awarded to white people (Wilson, 2016) and only 3.8% of tenured or tenure track positions at the top 100 geoscience departments are held by faculty of color (Bernard & Cooperdock 2018). There are multiple reasons at the heart of this issue including the lack of representation which can fuel stereotype threats and imposter syndromes.

This project aimed to take actions that would help grow the diversity, equity, and accessibility within Geneseo’s GSCI department. This …

369— Comparison Of Wilkins Creek And Spring Creek Stream Chemistries, Victoria Taylor, Brooke Stickles

GREAT Day Posters

A stream's chemistry and health can be affected by seasonal fluctuations, environment and land usage. Wilkins Creek and Spring Creek are tributaries of Conesus Lake located in Livonia, New York. The two streams flow through varying terrain. Wilkins Creek flows predominantly through wetlands, residential and agricultural land. Spring Creek runs through the village of Livonia. The different environments may result in altered stream chemistries. An analysis of water chemistry was taken at different points in Wilkins Creek and Spring Creek before they converged. Alkalinity, chloride, oxygen, pH, conductivity, and water temperature were measured at each location. Samples were collected and …

328— Watershed Analysis Of The Eberswalde Delta (Early Hesperian), Mars, Jason Mueller

GREAT Day Posters

The Eberswalde Delta is an impressive Martian fluvial-deltaic system. It is fed by an Earth-like dendritic river system that indicates past precipitation. This study is designed to quantify controls on network morphometry and its past extent using Esri ArcMap. Upstream contributing area (km2), elevation (m), depth (m), width (m), W/D ratio, and slope (-m/m), were calculated and graphed by stream order and long profiles. Width/depth increased with distance downstream but fluctuated in the fifth-order channel that was convex with knickpoints. This implied the system was youthful. The fourth-order stream was concave and in a sedimentary sub-basin. This sub-basin may have …

006— Pretreatment And Fiber Content Analysis Of Cannabis Sativa, Dina Bu, Jessica Roggie, Sarah Schmidlin, Barnabas Gikonyo

GREAT Day Posters

Cannabis sativa commonly known as hemp is one of the fastest-growing plants whose refined products have immense commercial value. Various products include refined hemp such as: biofuels, biodegradable plastics, textiles, dietary supplements, paper, clothing, and much more. Hemp fibers are also used in construction and manufacturing applications by strengthening their composite products. Hemp is a high yielding, sustainable, and environmentally friendly crop due to its various qualities, and has the potential to yield valuable raw materials for a great number of applications. Our research evaluates the pretreatment of hemp as well as the comparative analysis of the fiber content thereof. …

344— Wetland Soil, Abigale O'Connor

GREAT Day Posters

This project classified soil properties in the vicinity of a wetland in Livonia, NY. Soil cores were collected from the margin of the wetland to observe the changes in the soil profiles across the transition between the wetland and the surrounding forested land. Soil horizons were delineated based on texture and color. Individual horizons were then correlated between the cores to capture spatial changes in thickness and texture. The core descriptions and the correlations were digitized to create an archive of correlations around the wetland for future reference. The distribution of wetland soil textures contributes to better understanding how water …