Matching Items (1,362)
Filtering by
- Creators: Computer Science and Engineering Program
Description
Using the GICS methodology, MSCI and S&P DJI develop sector indexes that are reconstituted quarterly. ETF investment managers have an incentive to reduce tracking error to their underlying indexes and will buy securities that have been recently included in a reconstitution. I conduct an event study on the price effects of firm inclusion into sector ETFs. I document the existence of pre-inclusion excess return in a sample of stocks added to sector ETFs.
ContributorsBeeter, Nicholas (Author) / Wahal, Sunil (Thesis director) / Pruitt, Seth (Committee member) / Barrett, The Honors College (Contributor) / Department of Finance (Contributor) / Department of Marketing (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 However, the company has an outdated process of keeping track of metrics and retaining information about the production metrics of each employee.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
Description
This project presents a high-performance implementation of Gotoh’s pairwise sequence alignment algorithm within the scikit-bio Python library, optimized using Cython for computational efficiency. Designed to address the limitations of scikit-bio’s original pure Python alignment function, the new implementation achieves substantial performance gains—exceeding a 2800x speedup on longer sequences—while maintaining accuracy and full compatibility with the existing scikit-bio API. Emphasis was placed on modular code structure, enabling ease of use, maintainability, and seamless substitution between global and local alignment modes. Benchmarking against BioPython, BioTite, and the prior scikit-bio implementation confirms the new algorithm’s competitive runtime and practical value for real-world bioinformatics workflows. This work demonstrates the potential for integrating low-level performance enhancements within high-level, user-friendly scientific computing environments.
ContributorsAzom, Raeed (Author) / Zhu, Qiyun (Thesis director) / Aton, Matthew (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
The Intel Collaborative Thesis Program matches teams of selected Business Honors students with managers from Intel Corporation.** The teams work together on a topic identified by the Intel managers, who serve as coaches throughout the process. Based on the topic chosen and with input from the team coaches, the thesis will first define the problem to be addressed and outline why Intel is interested in this problem or opportunity. The thesis will then present an analysis that the team believes will help Intel successfully address the problem or opportunity that was defined. Our project evaluated strategic options for Intel to maximize long-term shareholder value in light of declining revenue, margin compression, and reduced market valuation. We specifically examined whether Intel should continue operating as an integrated device manufacturer (IDM), spin off its foundry and products businesses, or pursue alternative strategies. Through in-depth financial analysis, benchmarking against competitors such as AMD and TSMC, and assessment of Intel’s IDM 2.0 strategy, we analyzed the potential impacts on efficiency, valuation, and capital generation. Our recommendation outlines a path forward that we believe best positions Intel for sustainable growth and increased shareholder returns.
ContributorsWright, James (Author) / Moore, Mason (Co-author) / Ahmed, Selma (Co-author) / Naik, Shivani (Co-author) / Simonson, Mark (Thesis director) / Hertzel, Mike (Committee member) / Barrett, The Honors College (Contributor) / Department of Finance (Contributor) / School of Accountancy (Contributor) / Department of Information Systems (Contributor)
Created2025-05
Description
The evolution of digital commerce has fundamentally reshaped the landscape of global
financial transactions, establishing an ecosystem where speed, operational efficiency, and user convenience are the highest priority. This transformation, driven by the adoption of
internet-based platforms and mobile technologies, has enabled unprecedented levels of
accessibility and connectivity in economic exchanges. However, this rapid expansion has caused a significant escalation in payment fraud, which exploits vulnerabilities inherent in contemporary transaction systems. Historically, platforms such as Stripe Radar have adopted a centralized approach to fraud prevention, aggregating extensive datasets—including credit card details, behavioral analytics, and device fingerprints—to identify and mitigate fraudulent activities. This methodology, while effective in certain contexts, introduces substantial risks due to the concentration of sensitive information within a single architectural framework. Such
centralization renders these systems prime targets for cyberattacks, as evidenced by high-profile breaches that have compromised millions of users’ personal and financial data.
ContributorsGundala, Revanth (Author) / Boscovic, Dragan (Thesis director) / Bazzi, Rida (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / Dean, W.P. Carey School of Business (Contributor) / Economics Program in CLAS (Contributor)
Created2025-05
Description
Under the supervision of The Difference Engine, specifically the Women’s Power and Influence Index (the Index), this team (Amira Abboud, Natalie Alexander, Ava Bianco, and Sophia Bowers) created a video training series to educate women on the methodology and criteria of the Index. To test the efficacy of this video training series, the team designed a survey asking respondents to answer several questions about the topics within some of the videos in this training series. Despite a self-selecting sample, the team received statistically significant results, meaning their videos significantly impacted the respondents’ awareness and understanding of the topics within the videos. The team hopes that their work on this thesis will contribute positively to the Index, and that their work will inspire gender equitable policy in the workplace.
ContributorsAbboud, Amira (Author) / Alexander, Natalie (Co-author) / Bowers, Sophia (Co-author) / Bianco, Ava (Co-author) / Zaffar, Ehsan (Thesis director) / Gel, Esma (Committee member) / Barrett, The Honors College (Contributor) / Department of Finance (Contributor) / School of Accountancy (Contributor)
Created2025-05
Description
GLOW delivers thoughtfully curated wellness and snack boxes designed to support student energy, focus, and overall well-being. Each box features a balanced mix of functional items, including Liquid I.V. with caffeine, premium trail mix, and a daily supplement pack. The premium version also includes Wiff aromatherapy as a calming boost. These selections are carefully chosen to help students reset and recharge throughout the week. Priced at $15, GLOW offers an accessible and budget-friendly solution for students looking to prioritize their health. Sourcing and packaging decisions are made with care, reflecting a strong commitment to sustainability and responsible choices.
ContributorsNelson, Spencer (Author) / Israel, Anoushka (Co-author) / DeMartin, Claire (Co-author) / Greenhill, Dylan (Co-author) / Vanderwey, Lina (Co-author) / Byrne, Jared (Thesis director) / Howell, Travis (Committee member) / Barrett, The Honors College (Contributor) / WPC Graduate Programs (Contributor) / Department of Finance (Contributor) / School of Accountancy (Contributor)
Created2025-05
Description
Machine learning has increasingly played a pivotal role in societal decision-making. In such contexts, ensuring the fairness of models becomes critically important. Unfortunately, prior studies have shown that without intervention, models often inherit and even amplify biases present in the datasets. Existing group fairness metrics, such as true positive rate parity and equalized odds, primarily focus on the relationship between model predictions and ground truth labels stratified by sensitive attributes. However, most existing notions of fairness overlook whether the underlying rationale of a model’s decision-making process varies across different subgroups. To fill this gap, we propose a novel metric, FIDSHAP, which evaluates fairness through explainability by quantifying discrepancies in the model’s decision rationales across groups. Subsequent experiments and optimization procedures validate the effectiveness of this metric and underscore the potential of addressing fairness from the perspective of explainability.
ContributorsQiu, Ziyue (Author) / Choi, YooJung (Thesis director) / De Luca, Gennaro (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05