Matching Items (746)
Filtering by
- Creators: Computer Science and Engineering Program
Description
While introductory machine learning education emphasizes both theoretical foundations and programming syntax, the ability to confidently navigate data workflows and make informed decisions is typically developed through repeated, project-based experience. Interest in machine learning education is expanding simultaneously with the widespread availability of generative artificial intelligence (GAI) tools, capable of automating data analysis, decision-making, and code generation. While convenient, GAI tools may risk undermining students’ development of essential reasoning skills when used without a solid understanding of the underlying concepts. This issue may be particularly pressing for machine learning students attempting to conduct a machine learning workflow, already balancing the dual demands of learning programming and machine learning theory.
DataPylot is an interactive application designed to support early learners in machine learning by fostering strategic intuition around effectively applying machine learning techniques to data. The tool provides a structured interface that enables users to upload a dataset of their choice and apply common supervised machine learning tasks step-by-step. At each step, users make explicit choices through a graphical interface, and the application deterministically maps these selections into formatted Python code ready for execution. DataPylot supports code generation for many typical tasks within a machine learning workflow, including importing a raw dataset, applying exploration and preprocessing techniques, and training and evaluating machine learning models. This approach places the reasoning and decision-making process in the hands of the learner while reducing programming barriers and minimizing common errors.
To evaluate DataPylot’s educational value, a user study was conducted with Arizona State University students who had prior experience with machine learning. Participants successfully completed a guided machine learning project using only code generated within DataPylot to explore, preprocess, and model a provided dataset. Afterwards, participants completed a questionnaire assessing perceived challenges in learning machine learning, the ease of use and value of using the tool, and its comparative utility relative to GAI tools. Results indicated that participants found DataPylot useful for applying machine learning and supporting independent reasoning, particularly among lower-experience learners. These findings provide support for the potential of structured, interactive, and deterministic tools like DataPylot to foster conceptual understanding and strategic intuition surrounding applying machine learning to data through guided hands-on engagement.
ContributorsRice Nulty, Seth (Author) / Chavez Echeagaray, Maria Elena (Thesis director) / Zhu, Qiyun (Committee member) / Barrett, The Honors College (Contributor) / Department of Psychology (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
This project introduces a new platform built on Kubernetes that makes it easy to run small, efficient programs (serverless functions) using a technology called WebAssembly. It aims to improve upon existing systems that don't work well with WebAssembly. Users can simply upload their code through a web interface, and the platform automatically handles all the complex steps needed to package and run it on Kubernetes using standard industry approaches. The goal is to provide developers with a straightforward way to build and deploy WebAssembly-based serverless functions in a modern cloud environment.
ContributorsMykhaylov, Michael (Author) / Bazzi, Rida (Thesis director) / Cook, Corey (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2025-05
Description
Canada’s dynamic immigration system attracts thousands of newcomers each year, yet many immigrants face persistent challenges in achieving full economic, social, and cultural integration. This thesis addresses these challenges through the design and development of a mobile application aimed at supporting immigrant settlement and integration. The application functions as a personalized tool that calculates the gap between an immigrant’s current status, based on key metrics such as immigration status, student history, skill metrics, employment history, and desired integration goals. By leveraging user input and publicly available settlement benchmarks, the app provides targeted, actionable recommendations tailored to each individual’s circumstances. The goal of the app is to empower users with clear, data-driven pathways toward successful integration. The thesis includes a review of integration theory, an analysis of current settlement support infrastructures in Canada, and a usability evaluation of the app among recent immigrants. The findings suggest that digital tools like this app can play a vital role in bridging information gaps, promoting self-advocacy, and enhancing long-term settlement outcomes.
ContributorsCantoni, Emma (Author) / Osburn, Steven (Thesis director) / Khan, Saad (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
A custom physical-layer networking protocol was created to enable file transfer between two Ultra-Wideband boards. This is an exciting advancement because Ultra-Wideband isn't widely used in the consumer space for data transfer. The newly defined protocol allows Ultra-Wideband to be utilized for data transfer on smaller, more affordable hardware than its industrial counterparts.
ContributorsSweeney, Zachary (Author) / Osburn, Steven (Thesis director) / Kim, Hokeun (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / The Sidney Poitier New American Film School (Contributor)
Created2025-05
Description
Selective radiative coolers (SRCs) were incorporated into a high-resolution WRF simulation of Arizona for May to June 2024 to assess urban cooling efficacy. SRCs delivered 0.5 to 0.9 °C cooling in Tucson. Model validation against urban weather stations showed reasonable levels of error in the model, supporting reliability and comparison. Overall, this marginal cooling underscores the need for broader strategies to be employed to combat urban heat in Arizona
ContributorsSwart, Ryan (Author) / Georgescu, Matei (Thesis director) / Moustaoui, Mohamed (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / School of Mathematical and Statistical Sciences (Contributor)
Created2025-05
Description
In our preliminary coding classes, there was little clarity on how the technical skills we were learning would eventually play a role in our professional lives. This inspired Cypher Workshops, a project focused on designing an exploratory curriculum covering the principles of web development and industry-standard tools. We believe that exposing students to the large-scale applications of each skill first can help them flourish within their chosen interest, or even get creative around the applications of what they are learning. We created a curriculum tailored for high schoolers–covering programming with React (JavaScript) and Flask (Python), designing in Figma, and using Git and GitHub for version control and collaboration.
To test the curriculum, we organized and hosted a 6-week long workshop series at a local high school. The workshops provided students with an opportunity to experiment with full-stack development and tools in a stress-free environment where they could learn to code, ask questions, and develop their programming skills without the pressure of grades or assessments. Ultimately, students ended the workshop with impressive projects and deepened awareness of their interests.
ContributorsCage, Quinn (Author) / Mittal, Sanya (Co-author) / Osburn, Steven (Thesis director) / Cherilla, Jennifer (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
In our preliminary coding classes, there was little clarity on how the technical skills we were learning would eventually play a role in our professional lives. This inspired Cypher Workshops, a project focused on designing an exploratory curriculum covering the principles of web development and industry-standard tools. We believe that exposing students to the large-scale applications of each skill first can help them flourish within their chosen interest, or even get creative around the applications of what they are learning. We created a curriculum tailored for high schoolers–covering programming with React (JavaScript) and Flask (Python), designing in Figma, and using Git and GitHub for version control and collaboration.
To test the curriculum, we organized and hosted a 6-week long workshop series at a local high school. The workshops provided students with an opportunity to experiment with full-stack development and tools in a stress-free environment where they could learn to code, ask questions, and develop their programming skills without the pressure of grades or assessments. Ultimately, students ended the workshop with impressive projects and deepened awareness of their interests.
ContributorsMittal, Sanya (Author) / Cage, Quinn (Co-author) / Osburn, Steven (Thesis director) / Cherilla, Jennifer (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
The App at Barrett (TAAB) is a mobile application designed for iOS and Android that consolidates key Barrett information into a single, accessible platform. It brings together event information, campus details (water locations, parking, bathrooms, etc), and the Barrett store into one platform. This enables students, staff, and visitors to easily digest the current day-to-day operations of Barrett and increases accessibility to the information already present.
ContributorsJha, Arvin (Author) / Fette, Donald (Thesis director) / Frankenfield, Angela (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
Falcon Engineering Corporation is a computer numerical control, textiles, and slings manufacturer. One of the company’s specialties is parachute manufacturing for both the military and civilian sectors. With many high profile clients such as Cirrus1 (a plane manufacturer), quality control is an extremely important domain. However, the company has an outdated process of keeping track of metrics and retaining information about the production metrics of each employee. Each employee has to keep track of the number of parachutes they assemble, detail rework information, and retain different metrics, including the amount of parachutes passed and failed. In the past, this has been done via paper. The work done as part of this thesis aims to modernize the quality assurance process by creating a managerial and mobile system, containing quality control forms, production metrics of each employee, and a way to display current trends within the employee production landscape. Based on the application created for Falcon Engineering, a script was distributed to different employees, walking them through the different processes the system can partake in. When surveying management about the usefulness of this software, they gave the overall software a 4.67 out of 5 stars, rating different aspects of the user interface, such as pass or fail bar chart, overall production, and the digitized rework form. The responses exemplify the usefulness of the application, with the main beneficiaries being the textile assembly employees, who now have a streamlined way of documenting quality control, and management, who now are able to see the quantity passed and quantity failed by each factory sewer in real time.
ContributorsKlonaris, Nathan (Author) / Chavez Echeagaray, Maria Elena (Thesis director) / Werner, Sean (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
Accurate drone localization in urban environments remains a challenge due to GPS signal blockage, multipath interference, and unreliable vertical positioning caused by dense architectural structures. This research investigates an alternative approach using Immersal’s visual positioning system (VPS) to enable image-based localization without relying on simultaneous localization and mapping (SLAM) or ARFoundation for mobile devices. By adapting the Immersal pipeline to accept external camera input, this work simulates a drone-based setup using webcam footage and estimates focal parameters to support localization. While real drone deployment is outside the project scope, the resulting software provides a foundation for future integration with drone hardware by identifying the necessary sensor data for visual localization and connecting the necessary pipeline data. This approach lays the groundwork for infrastructure-free navigation in GPS-degraded urban environments, and the system has successfully demonstrated the ability to generate maps and extract camera poses using custom captured images run through Immersal. This was validated through webcam-based tests and offline drone footage, where Immersal returned consistent pose estimates and successfully built .ply-format spatial maps using synchronized image-pose data.
ContributorsColyar, Adam (Author) / Chavez, Maria (Thesis director) / Baillot, Yohan (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05