Collection

2024 ASEE North East Section

Authors

Christina Cao, .; Danushka Bandara, Fairfield University

diagnostics, and criminal justice outcomes. To address these issues, we propose a novel approach to evaluate bias in LLMs using metrics developed by Stereoset [1]. Our experiments involve evaluating several proprietary and open-source LLMs (GPT4, GEMINI PRO, OPENCHAT, LLAMA) for stereotypical bias and examining the attributes that influence bias. We used a selected 100 prompts from the stereoset dataset to query the LLMs via their respective APIs. The re- sults were evaluated using the language modeling score, stereotype score and the combination iCAT[1] score. In particular, open source LLMs showed higher levels of bias in handling stereotypes than proprietary LLMs (40% average stereotype score for

Collection

2024 ASEE North East Section

Authors

Daniel Blessner, Pennsylvania State University, Wilkes-Barre Campus

studentscan read the x-intercepts from the equation. The student graphs the polynomialwith the associated End Behavior and once again the solution is visually obtained.To simplify the procedure only linear factors with multiplicity one will be used.This is a work in progress (WIP) paper that aims to ensure that mathematicallyunder prepared engineering students succeed in their calculus series courses.Plans for a survey for the effectiveness of this new approach are planned for futuresemesters. Currently this new approach is available for anyone who wishes toimprove and reinforce basic graphing techniques. 1The ability to understand and interrupt graphs is critical for technical fields such asengineering [1

Collection

2024 ASEE North East Section

Authors

Gonca Altuger-Genc, State University of New York, Farmingdale State College; Akin Tatoglu, University of Hartford

educational setting has its own unique set of challenges both for facultyand students. From faculty’s perspective, it can easily be categorized as a tool students can usefor plagiarism. From students’ perspective it can be trusted to generate meaningful and relevantanswers. In addition, both students and faculty deal with the unknown as this is a new technology,there is hesitation and worry that comes with it as well.Our study offers a multi-faceted multi-level approach: Level 1, Preliminary Assessment: • Conduct a pre-experience survey to evaluate students' familiarity and attitudes toward AI-based tools, particularly ChatGPT. • Utilize a 6-question survey employing a 5-point Likert scale to capture nuanced responses

Collection

2024 ASEE North East Section

Authors

Tahmid Latif, Wentworth Institute of Technology

considered mundane orunsafe for humans. While the more traditional utility of robots lies in industrial applications,demand for robots has increased in areas like medicine [1], agriculture [2-3], and search-and-rescue [4-6] with the potential of becoming the next home “appliance.” As such, the Intro toRobotics course aimed to make engineering students capable users and designers in thisinterdisciplinary field. This paper details the strategic design of a linear course to establish orincrease key concepts and knowledge of robotic theory in students through a series of learningmodules and accompanying hands-on experiments.The area of robotics can be considered to lie at the intersection of topics like linear algebra,electronic circuits, data fusion

Collection

2024 ASEE North East Section

Authors

Abdullah Aldwean, University of Bridgeport ; Dan Tenney, University of Bridgeport

Tagged Topics

Diversity

services, which include enhancing patientexperience, improving population health, reducing costs, and improving the work life ofhealthcare providers [1]. Emerging technologies such as Artificial intelligence (AI) have thepotential to transform the healthcare sector by improving patient experience, increasingoperational efficiency, and advancing medical research. Researchers in the medical field widelyacknowledge the potential role of AI applications in revolutionize traditional healthcare modelsand shift the service toward data-driven, patient-focused care [2],[3]. However, among many AIapplications, the Large Language Model (LLM) has been recognized as one of the mostpromising AI applications in the healthcare sector [4]. Unlike many traditional

Collection

2024 ASEE North East Section

Authors

Djedjiga Belfadel, Fairfield University; Isaac Macwan, Fairfield University; Naser Haghbin, Fairfield University; John F Drazan, Fairfield University

forEngineering and Technology (ABET), the course ensures that students are well-equipped to confront themultifaceted challenges of the engineering profession with confidence and competence. This paper servesas a testament to the efficacy of the Fundamentals of Engineering course in nurturing well-roundedengineers who are poised to make meaningful contributions to their fields upon graduation.Keywords: Fundamentals of Engineering, experiential learning, problem-solving, professionalism,MATLAB, Arduino, Solid works. 1. IntroductionEngineering design is defined as the communication of a set of rational decisions obtained with creativeproblem solving for achieving certain stated objectives within prescribed constraints. The role of design inan

Collection

2024 ASEE North East Section

Authors

Abhishek Kumar, Wentworth Institute of Technology

are self paced free online course offered byMathworks. In my interaction with Industry which recruits Wentworth graduates a frequentadvice is that students of all majors should be comfortable with Programming and Simulation.Many papers discuss how Simulink can be used effectively in areas such as dynamic systems,[1],[3],[6] and mechanical systems,[2],[9] and [7].Model Description:In this section we have developed the Matlab and Simulink models for 3 problems 1) Spring Mass Damper System under free Response 2) Spring Mass Damper System under forced Response 3) Simulink Model for Damped system with Forced ResponseSpring Mass Damper System under free Response m = mass of

Collection

2024 ASEE North East Section

Authors

Robin Chataut, Quinnipiac University; YUSUF USMAN, Quinnipiac University; Frederick Scholl, Quinnipiac University

protection [1].IdentityIQ's report underscores the sheer volume and variety of cyberattacks in 2023, noting a14% increase in data breaches over the previous record, affecting over 66 million individuals [2].The top ten data breaches over the last twelve months further illustrate the diverse nature ofcyber threats, ranging from massive data leaks to targeted attacks on specific organizations, asshown in Table 1. These incidents have collectively affected billions of accounts and highlightedvulnerabilities across various sectors, necessitating the implementation of more robust securitymeasures such as multifactor authentication [3].Table 1: Top ten data breaches over the last 12 months [3] Rank Name of the Breach/Attack 1 Russian Web Hosting Data