14th IEEE Integrated STEM Education Conference — 9 AM - 5 PM EDT, Saturday, March 9

Onsite Venue - McDonnell and Jadwin Halls, Princeton University, NJ - Virtual Attendees - Enter Zoom Room

Works in Progress Papers

Session WIP-01

Works-In-Progress 01 - virtual

Conference
2:30 PM — 3:15 PM EST
Local
Mar 9 Sat, 2:30 PM — 3:15 PM EST

Exploring STEM Career Competencies with the Assistance of Generative AI

Nurten Karacan Ozdemir (Boston University & Hacettepe University, USA); Chong Myung Park and Scott Solberg (Boston University, USA)

0
This paper presents preliminary findings of an analysis of STEM career competencies and skills using generative AI. Using PRISMA methodology, 65 research papers and 38 reports focused on STEM career competencies and skills met inclusion and exclusion criteria. The identified papers and reports were analyzed using Claude AI which adopts Large Language Models (LLMs), advanced artificial intelligence systems trained on large amounts of text data. Some research papers and reports were categorized by their specific focus, such as K-12 studies, studies regarding diverse STEM careers, and fabrication-lab and maker-space studies, and then they were analyzed separately. Through a combination of the skills and competencies that emerged from each analytic step, a final analysis was carried out. Being aware of the limitations of LLMs, the authors verified the themes based on the literature and further verification is in process with topic modelling analysis with LDA in R. The collective findings indicated six key STEM skills and competencies: technical skills, problem solving and critical thinking, research skills, soft skills, lifelong learning and adaptability, and other competencies. Specific themes by focus category include creativity and innovation, communication and collaboration from the fab-lab and makerspace research category, mathematical and analytical skills, ethical and social responsibility, passion, and dedication from research into diverse STEM careers, self-efficacy from K-12 STEM research, and more social and emotional learning (SEL) skills from STEM and employability reports. The findings indicated the importance of foundational technical skills as well as SEL skills for STEM career identity.
Speaker
Speaker biography is not available.

The Penetration of Generative AI in Higher Education: A Survey

Mario E. S. Simaremare and Chandro Pardede (Institut Teknologi Del, Indonesia); Irma N Tampubolon (Del Institute of Technology, Indonesia); Daniel Andres Simangunsong (Institut Teknologi del, Indonesia); Putri Esrahana Manurung (Del Institute of Technology, Indonesia)

0
The global teacher shortage crisis is a serious challenge. The crisis also rises in Indonesia, where the problem extends to unequal teaching quality and learning facilities. This situation seriously threatens the Indonesian 2045 vision of being a developed country with knowledgeable human resources. The advancement of AI brings opportunities to address the challenges. In recent years, there has been a wave of generative AI (GenAI) technologies and their adoption in education. However, there is no research on the penetration of such technology in our learning environment. In this paper, we investigate the use of GenAI by students in a higher education setting. We surveyed 1,157 students of Institut Teknologi Del in bachelor's and associate's engineering programs. IT Del is a small private university in the western part of Indonesia. Our result shows that most students are well aware of GenAI technologies (70.96%) and have used them to support their learning (98.96%). The top five most used GenAI tools are GitHub Copilot, OpenAI ChatGPT, Codex, Grammarly, and ChatPDF. GenAI is already part of the learning process. We believe that sooner or later, GenAI will be one of many deciding factors in our future education systems, and we must be ready to adapt to it.
Speaker
Speaker biography is not available.

Unify and Broaden STEM Outreach Activities through Simulator Interfaces for Wireless Sensor Networks

Ramakrishnan Sundaram (Gannon University, USA)

0
This paper describes the setup of the platform to integrate the physical wireless sensor network with real-time network simulation tools thereby creating the unified environment for STEM outreach activities. The physical wireless sensor network comprises WiFi modules, configured in a grid, to transmit and receive radio frequency signal data. The real-time network simulation tools interface with the WiFi modules in the physical environment and enable the signal data from the physical sensor network to be assigned to virtual sensor nodes in the simulation environment. In this manner, the physical network is replicated on the simulator and the creation of the virtual sensor network facilitates the modeling of the actual sensor grid. CupCarbon is the wireless sensor network design and simulation tool used to interface with the WiFi modules used in the physical wireless sensor network. The paper illustrates STEM outreach activities which highlight the various aspects of the unification process.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-02

Works-In-Progress 02 - Virtual

Conference
2:30 PM — 3:15 PM EST
Local
Mar 9 Sat, 2:30 PM — 3:15 PM EST

Mastering NoSQL Through SQL Foundations

Xavier Amparo and Ching-yu Huang (Kean University, USA)

0
Database management systems are a cornerstone of software engineering projects. These systems are often run using MySQL which is often used for relational and structured data or NoSQL which is for unstructured or semi-structured data. NoSQL is often used for real-time analysis of big data, content management systems, and document management systems. Many companies such as Google or Facebook have their own NoSQL databases that they use for their businesses, with BigTable and Cassandra being the examples for the aforementioned corporate entities. MongoDB is an open-source document database classified as a NoSQL database and is the platform used for NoSQL in this study. This study has the purpose of making a comparison between NoSQL and MySQL. This study is also to understand the best way to teach NoSQL in classrooms to students as right now, NoSQL is not typically taught as its own topic or course subject in classes, with MySQL taking priority. A comparison between both MySQL and NoSQL can help with examining the effects of how the prior knowledge from taking a class that utilizes MySQL can be used to learn NoSQL. An accessible method for learning the management system exists in the form of MongoDB, an accessible platform for students to interact with and learn the language.
Speaker
Speaker biography is not available.

Space Exploration Museum with Virtual Reality

Joshua Maddy, Neil Loftus, Hannah Vitalos, Haleigh Zerkle and Husnu S Narman (Marshall University, USA)

0
This research paper documents the development and ongoing testing of a Virtual Reality (VR) modern museum experience for people of all ages, especially those who face financial or physical barriers to visiting traditional institutions. This paper is a continuation of the Metaphysical Exhibition project, which aims to provide a cost-effective and easily accessible alternative to physical museums by utilizing VR technology that can be accessed with a one-time purchase of hardware and free software. The virtual museum is portable and can be experienced anywhere. Our focus is on developing a VR museum about space exploration and assessing its impact on users. Through our planned study, we aim to answer three key questions: (i) How does the VR museum experience compare to that of a physical museum? (ii) What is the level of interest and engagement in self-guided VR content for educational purposes, from both students and teachers? (iii) How can the VR museum experience be enhanced to optimize learning outcomes and user satisfaction?
Speaker
Speaker biography is not available.

Analysis of FDA Adverse Event Reporting System Based on Toolformer

Bo Wang (University of Southern California, USA)

0
The paper introduces Toolformer, a model designed to enhance the performance of large language models, particularly on the FDA dataset Federal Adverse Event Reporting System (FAERS). Large language models excel in various domains but can be unreliable for simple tasks. Toolformer addresses this by leveraging multiple APIs to improve accuracy, using a dataset with potential API calls generated by a language model. It selects valid tokens through self-supervised loss and fine-tunes based on inferred API calls, improving its ability to choose tools. Toolformer, when applied to the FAERS dataset, demonstrates reliability and minimizes biases in outputs, enhancing credibility in healthcare. The model implies benefit in addressing concerns about the FDA's post-market surveillance focus. Despite limitations, such as the inability to chain outputs between tools, Toolformer significantly improves the professionalism of large language models in healthcare applications.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-03

Works-In-Progress 03 - Virtual

Conference
2:30 PM — 5:00 PM EST
Local
Mar 9 Sat, 2:30 PM — 5:00 PM EST

Curriculum Development for UAV Cybersecurity

Jiawei Yuan (University of Massachusetts Dartmouth, USA)

0
Echoing the pressing demand to promote education and scientific research in cybersecurity and UAVs, this project seeks to improve UAV cybersecurity education towards a broad audience through the development of curriculum materials. The curriculum materials will help improve the education of cybersecurity in UAVs in terms of curricula, student learning, hands-on practice, as well as faculty collaboration. This project develops curriculum materials using a modularized manner. These course modules can not only be combined as a complete course but also integrated into existing cybersecurity or UAS curricula as a special topic section. The threat-driven approach is adopted in the design and development of course modules, which provides a destructive thinking manner that matches inquiry-based learning.
Speaker
Speaker biography is not available.

Remote Sensing for Year-Round Biodiversity Estimates: In Support of Monitoring During Extreme Events

Katharine McDaniel (Johns Hopkins University Applied Physics Lab, USA); Krista Rand (University of Virginia, USA)

0
This paper contains a discussion of application of multi-season data in forest biodiversity estimates from remote sensing. Current approaches rely only on summertime data, which is insufficient for multi-season, accurate, real-time assessment. Given the dynamic pace of development, wildfire activity, and other landscape-scale changes to ecosystems, methods that are sufficiently robust for year-round monitoring applications are critical. Methods surveyed include Rao's Q Diversity Index, coefficient of variation, NDVI, enhanced vegetation index, tasseled cap transformation, and Shannon's H Index. Remote sensing data sources include LIDAR and open-source satellite data. Excluding measures of greenness in situations where it is inappropriate for the mix of species and climate, this comparison will investigate the efficacy of tree species diversity indicators at seasonal and annual time scales, in both deciduous and boreal forest contexts. Different classifications of forests will respond uniquely to each approach, and the intricacies of these results are recorded to ensure appropriate implementation in the future. The reliability of each indicator is determined through comparison to previously verified values – either in-situ derived data or remote sensing estimates from established combinations of methods and seasons. Ultimately, the objective is to understand whether current approaches suffice to characterize canopy-level biodiversity throughout the year, propose potential modifications to improve characterization outside the summer season, and introduce other practical or theoretical considerations that may be relevant to ecologists and other biodiversity researchers and professionals.
Speaker
Speaker biography is not available.

Building Cybersecurity Mindset through Continuous Cross-module Learning

Jiawei Yuan (University of Massachusetts Dartmouth, USA); Yanyan Li (California State University San Marcos, USA)

0
Recent years have witnessed a strong demand for cybersecurity professionals, especially considering the remarkable projected growth rate of cybersecurity positions for the next ten years. Filling such a workforce demand is not simply preparing dedicated cybersecurity students. More importantly, it is essential to establish a cybersecurity mindset for computer science students in general, because they will contribute to the majority of the cybersecurity-related workforce in the near future. This project aims to promote cybersecurity education for undergraduate computer science students in general and prepare a future cybersecurity workforce. Instead of heavily relying on dedicated cybersecurity courses in most existing undergraduate cybersecurity education, this project proposes the concept of cross-module learning that leverages non-cybersecurity computer sciences courses at different levels to continuously instill security concepts and skills to train students from the early stage of undergraduate learning. To enable effective cross-module learning, we identify appropriate computer science courses to integrate cybersecurity modules, investigate effective integration strategies, and explore potential linkages among different courses and modules.
Speaker
Speaker biography is not available.

Evaluating the Inclusion of Strategic Management Education in Technology Entrepreneurship

Frank Washko (Saint Martin's University, USA); William Edwards (Kettering University, USA); Leslie Washko (Saint Martin's University, USA); Shahlaa Al Wakeel (Saint Martins University, USA)

0
This work studies to what extent strategic management education enhances the overall understanding of students in a technology entrepreneurship class. Past students who completed the class did well building business plans and marketing plans, but lacked an overall understanding of strategy and business models in their projects. A subsequent class included modules on these topics. The results were surveyed before and after the inclusion of the material. Positive results were noted in the students' projects, and additional insights were gathered from the students. Students demonstrated a better understanding of the relationship between business strategy, business models, and technology. They further identified areas that would further enhance their understanding of technology entrepreneurship.
Speaker
Speaker biography is not available.

A Framework for Introductory Data Science Experiences for Non-Computer Science Majors

Rebecca Lowe and Adrienne Smith (Cynosure Consulting, USA); RN Uma (North Carolina Central University, USA); Sambit Bhattacharya (Fayetteville State University, USA); Debzani Deb (Winston-salem State University, USA); Alade Tokuta (North Carolina Central University, USA); Muztaba Fuad (Winston-Salem State University, USA)

0
The surge in data science courses highlights an educational shift towards incorporating data skills as a fundamental component of a well-rounded academic curriculum. This growth indicates a recognition of the critical role that data plays in shaping our understanding of complex issues in today's data-driven society. It also underscores the need for data science education to be accessible, representative, and tailored to a wide array of learners and professionals. The work to date to establish a data science foundational knowledge framework represents a pivotal step in formalizing and enhancing data science education. Yet, these frameworks have been largely designed for those pursuing data science careers. The field is in need of a data science framework that focuses on the essential introductory knowledge and skills for non-CS students to build a solid foundation in data science. This paper provides a description of the literature review process and experiences the research team has drawn from to develop such a framework. It also positions this framework for future research in studying effectiveness and alignment to the K-12 space.
Speaker
Speaker biography is not available.

Beyond Subjectivity: Exploring Factors Impacting Course and Teaching Evaluation

Zhizhen Chen (The Chinese University of Hong Kong Shenzhen, China); Ruizhi Liao (The Chinese University of Hong Kong, Shenzhen, China); Ao Zhang (The Chinese University of Hong Kong Shenzhen, China)

1
Course and Teaching Evaluation (CTE) is a widely used method for assessing teaching quality in educational institutions. While it is intended to provide instructors with feedback that can help them improve their teaching, studies have shown that subjective factors, such as students' expected grades, can affect evaluation scores. Moreover, instructors may alter the teaching content or grading strategy to improve their CTE scores, leading to grade inflation and low authenticity of evaluation results. To address these challenges, we tested all the factors that will affect CTE scores based on the available datasets and built a least square model for credibility validation. The model shows a high robustness in the testing set. By exploring alternative evaluation methods, this paper seeks to improve the reliability of teaching evaluation results, ultimately improving the quality of education.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-04

Works-In-Progress 04 - Virtual

Conference
2:30 PM — 3:15 PM EST
Local
Mar 9 Sat, 2:30 PM — 3:15 PM EST

Green Cement: Comparative Assessment of Portland Cement Alternatives to Reduce Carbon Dioxide Emissions

Kristine Won (USA)

0
The cement industry has steadily increased the amount of carbon dioxide emissions in the past several decades with other greenhouse gas emissions from global industries, contributing significantly to climate change. Alternative methods to creating Portland Cement, the most commonly used type of cement, are explored to evaluate how effective the alternatives would be in mitigating negatives effects of the cement industry while maintaining strength. In order to quantify the environmental impact of different concrete materials, a systematic analysis tool called the Life Cycle Assessment (LCA) was used to calculate the total carbon dioxide emissions. Alternatives such as replacement of limestone with geopolymer, Supplementary Cementitious Materials, different fuel types from biomass, and enzymatic self-healing cement significantly decrease the amount of carbon dioxide emissions. Further research on the alternatives, especially the enzymatic self-healing cement and its optimal conditions, will allow the cement industry to significantly reduce emissions contributing to global climate change.
Speaker
Speaker biography is not available.

Analysis of Organized Cyberbullying

Jieqi Wang (Nanyang Technological University, Singapore)

1
Organized cyberbullying has emerged as a critical societal issue with profound psychological and physical impacts on victims. This paper explores organized cyberbullying, delving into its importance, background information, and key theoretical underpinnings. Through a comprehensive literature review, the study examines global statistics, trends, and the academic and legal perspectives surrounding cyberbullying. Drawing from existing research, the paper highlights the power dynamics, deindividuation, anonymity, and technological mechanisms driving cyberbullying behaviors. The discussion includes analyses of potential solutions and interventions, considering the effectiveness of various strategies in combating organized cyberbullying. Additionally, the paper offers insights into future research directions and policy development to address this pervasive issue.
Speaker
Speaker biography is not available.

AI in Education: Crafting Policies for Tomorrow's Learning Landscape

Mehdi Roopaei (University of Wisconsin - Platteville, USA); Nasrin Dehbozorgi (Kennesaw State University, USA)

0
The educational sector is evolving, significantly influenced by artificial intelligence (AI). As AI becomes more prominent in society, there's a pressing need to reform educational systems to prepare students for an AI-driven future. Present models show a gap in preparing students for AI-rich job markets. Integrating AI into curricula can enhance learning and equip students with vital skills like critical thinking and problem-solving in an AI context. To effectively embed AI in education, comprehensive policies on ethical use, data privacy, curriculum integration, educator training, infrastructure, and legal compliance are crucial. A planned survey at the University of Wisconsin-Platteville and Kennesaw State University will explore these aspects. These institutions, known for their teaching and research focus respectively, will provide insights into varying educational approaches towards AI. The findings, aimed to be shared in a future article, will offer comparative perspectives on AI integration in different academic settings.
Speaker
Speaker biography is not available.

The QEd Project: developing quantum conceptualization in UConn's STEM curriculum

Jason N Hancock, Max Meynig, Brenna Petrelli, Lea Santos, Douglas Stewart, Diego Valente and Xian Wu (University of Connecticut, USA)

0
We report progress in reform of our teaching all calculus-based introductory physics classes taught by the Department of Physics at the University of Connecticut. In the period 2016-2021, we have successfully converted all calculus-based physics courses from traditional stadium lecture to an active learning environment, with original activities. In this context we will describe our efforts, metrics, and outcomes which demonstrate success in identifying an effective approach to teaching and creating a community of practice for educational excellence. The scope of students affected includes most science and all engineering majors. With our new teaching approach in hand, we describe a plan to turn our attention to the curriculum of these courses, in particular to introduce quantum concepts at the early stages of the STEM curriculum, integrating classical and quantum physics instruction to prepare the next generation of quantum-relevant workforce.
Speaker
Speaker biography is not available.

Automated Discourse Analysis System

Varun Gottam, Nasrin Dehbozorgi and Soon Lee (Kennesaw State University, USA)

1
Natural Language Processing is the process of making machines understand, interpret and respond in human language. The integration of NLP facilitates a multitude of educational processes including automated classes, interactive tutorials, and sophisticated feedback systems, all of which significantly enhance the efficacy of both teaching and learning. One such application is the requirement for analysis and discourse of discussions that happen between teachers and students especially in science classes. Educational researchers have developed a framework, CDAT (Classroom Discourse Analysis Tool), to code the type of discussions in the science classroom. Via this taxonomy, the educators manually tag the discourse patterns in the science classroom. Although the CDAT framework is instrumental in enriching the educational discourse, its manual application is both time- consuming and resource intensive. The automated system developed for this system, once implemented, can be deployed on the Internet and will be available to all the teachers across the country who can upload the transcripts. of their classroom discussions and get immediate and detailed feedback. By leveraging the power of AI, this system is designed to deliver promising analysis of classroom discourses, scientific patterns and student engagements. This integration of Artificial Intelligence in the education sector will be a pivotal point in improving the education technology in near future.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-05

Works-In-Progress 05 - Virtual

Conference
2:30 PM — 3:15 PM EST
Local
Mar 9 Sat, 2:30 PM — 3:15 PM EST

Comparative studies on student writing improvement-A cybernetics perspective

Yijun Li (Canterbury School, USA)

0
This study employed a cybernetics methodology to conduct competitive research on the possible advantages of offering a combination of two systems that provide students access to books classified by writing style and enables instructors to offer tailored comments on students' writing. First, the method may support students in gaining a more thorough understanding of the art of writing and developing into confident and competent authors by exposing them to various writing techniques and choosing books that match their writing style or those they want to mimic. The second approach, in which teachers may give individualized comments on students' writing from a cybernetics view, creates a feedback loop, which can assist students in better understanding their strengths and areas for growth and receiving focused advice on enhancing their writing. Combining the two approaches may improve pupils' writing abilities and self-assurance. The article opened with an overview of the significance of writing abilities and the difficulties students often experience in acquiring them. The study then examined the reevaluations of the link between reading and writing, the possible advantages of adopting a writing style-sorted book system, and the effect of individualized feedback on writing abilities. The research concluded by discussing these systems' implications for educators and recommendations for further investigation.
Speaker
Speaker biography is not available.

Unmanned Aerial Vehicle State Estimation using Ground Imagery in Global Positioning System Denied Environment

Madeline E Carter (Johns Hopkins Applied Physics Laboratory, USA)

0
The position and velocity of an unmanned aerial vehicle (UAV) can be estimated using ground features in successive camera images when the Global Positioning System (GPS) measurements are unavailable in a GPS-denied environment due to a variety of issues the vehicle could face while flying (ex: spoofing, obstacles, etc.). This paper compares the velocity estimate accuracy for different UAV cameras for different combinations of UAV altitude and velocity and then makes a recommendation for the ideal drone altitude and velocity to estimate the UAV position and speed from the image scene. This research aims to enhance UAV navigation capabilities in challenging environments, providing valuable insights for effective UAV operations in scenarios where GPS signals are unreliable or unavailable.
Speaker
Speaker biography is not available.

Integrating STEM using Design Thinking Wicked Problems: A Case Study from India

Richa Mishra (Institute of Technology, Nirma University, India)

0
This research work is a work-in-progress study. Design Thinking is a way of thinking which is used in engineering classrooms to bring human-centric solutions. The study has integrated STEM by amalgamating interdisciplinary subjects in STEM using the Design Thinking approach using a collaborative teaching method. The study aimed to address the research questions: Is it possible to use Design Thinking as an approach to integrate STEM? Is it possible to raise the empathy level of the students without compromising STEM learning? Can collaborative teaching give impetus of this integration? Empathy is a part of the Design Thinking process and a very valuable attribute for engineers. The study also investigated whether students become more empathetic due to the nature of this approach. The author has used wicked problems to reach the objective. The pilot study was done by recruiting first-year students of an engineering college. The intervention has used collaborative interdisciplinary STEM teaching. The results were assessed by 1. Working prototype and 2. Statistical analysis of the pre and post-test of the engineering knowledge, and 3. Analysis of IRI index test to check empathy. The study has yielded satisfactory results.
Speaker
Speaker biography is not available.

Toward a Framework for Providing Equitable Opportunities to Activate STEM Talent

Sydney Floryanzia (Johns Hopkins University Applied Physics Laboratory, USA); Aishwarya Jayabharathi (Johns Hopkins University, USA); Katherine-Ann Carr (Johns Hopkins University Applied Physics Laboratory, USA); Karla Gray-Roncal (Johns Hopkins University, USA); William Gray-Roncal (Johns Hopkins University Applied Physics Laboratory & Preparation Meets Opportunity Foundation, USA)

0
Throughout the United States, there are many different programs designed to support the development of Science, Technology, Engineering, and Math (STEM) talent to meet the national demand for a skilled workforce. These range from small, passion-driven volunteer programs to large, institutional programs with foundation or government funding. Though these initiatives differ in their target population, setting, and goals, they are united by a common mission to provide opportunities for students to explore and thrive in STEM careers. Developing new programs can be challenging, specifically in systematizing and measuring the effectiveness of these efforts - which can require skillsets beyond mentorship and education and reduce the time available to directly support student mentees. Here, we describe the beginnings of a systems-based approach to implement a solution based on ideas that have evolved from our engineering education research and practice over the past two decades. Integrating this approach can allow programs to more comprehensively consider and support diverse student needs in a variety of contexts, especially for trailblazing students in high school, college and graduate school - those with both high potential and significant barriers to opportunities in STEM.
Speaker
Speaker biography is not available.

Proposal of a Physics Course at Biomedical Engineering: Example of a Multitopic Content

Huber Nieto-Chaupis (Peru & Universidad Autónoma del Perú, Peru)

0
The program of biomedical engineering will play a noteworthy role at the present and subsequent decades due to the its crossing areas curricula in the medical field. This would suggest to design new and inspirative content with a possible twofold aim: professional and research. Thus, this paper proposes a physics course based at different topics but pedagogically engaged in order to keep the educational coherence. This proposal is illustrated through the modeling of neural synapse, demonstrating that an end-to-end content might be initialized with a basic example treated at the biological level. From experiences at 2022 and 2023, by which a physics course was done to engineering programs, a multi-content course has been designed whose aims are prospective 2th year students of biomedical engineering program. Survey and final qualifications have indicated that the present proposal might be a tangible option based at evidence.
Speaker
Speaker biography is not available.

CodeMentor AI: An Evolving Tool for Advanced Feedback in Engineering Code Review

Mourya Teja Kunuku (Kennesaw State University, USA)

0
In engineering education, the manual code review process is fraught with challenges such as time-intensive grading, inconsistent feedback, and inherent biases. To overcome these inefficiencies, this study introduces CodeMentor Large Language Model (LLM), an innovative automated code review method. CodeMentor LLM greatly improves the scalability and efficiency of code evaluation in educational environments by offering consistent, unbiased, and comprehensive feedback. Our discussion revolves around how well the system aligns with educational approaches. In contrast to conventional automated tools that only concentrate on error detection, CodeMentor LLM is designed to provide thorough feedback together with extra learning materials, optimization techniques, and best coding practices. This method promotes a deeper comprehension of coding principles by helping students throughout their learning path in addition to identifying faults. We also examine CodeMentor LLM's customizing features. Because of its adaptability, the system can support a wide range of programming languages and course levels, which accommodates different student learning preferences. Its scalability is also highlighted, demonstrating its suitability for use in various class sizes and educational settings. All things considered, this study introduces CodeMentor LLM as an potential innovative tool in engineering education, providing a more efficient and pedagogically sound substitute for conventional code review techniques. It offers a revolutionary method for teaching coding that improves student learning results and teaching effectiveness by utilizing cutting-edge AI.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-06

Works-In-Progress 06 - On-site

Conference
3:15 PM — 5:00 PM EST
Local
Mar 9 Sat, 3:15 PM — 5:00 PM EST

Interdisciplinary Synergy: Resources for Embedding Plugged and Unplugged Computer and Data Science Activities into the K-12 Curriculum

Ted Samaras (Franklin Township Public Schools, USA); Thomas J Marlowe (Seton Hall University, USA); Katherine Herbert, Vaibhav Anu, Sumi Hagiwara and Stefan Robila (Montclair State University, USA)

1
In the ever changing environment of education, the skills learned and applied in computer and data science (CADS) have expanded beyond the reach of siloed, single-subject teaching in the classroom. The need for interdisciplinary learning and constructs have increased as real-world experiences have become a necessity and time has become a scarcity. The progression toward this reality has been accelerated in a post-pandemic world. This paper reviews literature regarding this topic, looks at barriers impacting student learning, and provides resources to assist educators in weaving plugged, unplugged, and hybrid components into their interdisciplinary lessons to amplify the impact of CADS. It also provides a scenario for and a vision of learning if educators can utilize an interdisciplinary approach to their lesson and unit plans. The sample template provided focuses on a middle and high school education model where multiple teachers are responsible for teaching the same students in a team. The curriculum discussed in the template is based on topics covered in the New Jersey State Standards for eighth grade students, but is flexible enough to be molded for the needs of various grade levels and different state and local standards. By using multimodal techniques and generating greater synergy through interdisciplinary lesson constructs, more robust learning environments are created for both students and educators.
Speaker
Speaker biography is not available.

STEM Across Different Fields: Pandemic Modeling and Scaled Fermi Dirac Distribution Function

Michael Shur (Rensselaer Polytechnic Institute, USA)

0
Rapidly evolving technology fields will require the STEM students to develop problem-solving skills that could be applied to a wide variety of diverse fields and projects. This is possible to achieve because the same mathematical equations and functions find applications in diverse STEM areas. As shown in this paper, the key concepts of semiconductor physics such as the Fermi-Dirac distribution function and Vegard's Law (generalized to Scaled Fermi-Dirac function and Scaled Vegard's Law) are very useful in modeling and monitoring pandemics, such as COVID-19 or expected future mysterious pandemic Disease X discussed in DAVOS 24. In turn, these generalized concepts could find applications in modeling noise in radio frequency transistors and interpolating parameters of ternary compounds.
Speaker
Speaker biography is not available.

Fostering Diversity and Knowledge in Artificial Intelligence: An Inclusive Platform of Career Insights and Resources from Conversations with Leaders in the Field

Max Charney (USA)

0
Computer science has recently experienced a surge in popularity with the release of generative artificial intelligence models. Yet, while the popular imagination has been captured, evidence indicates that pre-college students, especially those from underrepresented populations, lack adequate education about careers in the field. Effective methods for gaining understanding include shadowing professionals or acquiring internships, yet these means are often not attainable or realistic for many students. Overall, this insufficient education can lead to future disadvantages, including future workforce disparities. To address these problems, Computer Science Next (CSNext) is an organization that was founded in early 2024 to convey insights from the current to the next generation of computer scientists. CSNext's first project will be a series of video interviews with leaders in artificial intelligence intended for a high school student audience. Video interviews will be released for free, and responses will be synthesized to share meaningful and insightful commonalities and responses. In addition, a list of the resources discussed in the interview will be cataloged alongside the videos and indexed for the site in its entirety. Particular attention will be focused on inclusivity, including underrepresented minorities, when selecting interview candidates.
Speaker
Speaker biography is not available.

Hands-on Project Oriented Cybersecurity Education for High School Students

Qiaoyan Yu, Sandeep Sunkavill, Mashrafi Kajol, Mohammad Monjur, Nishanth Chennagouni, Diliang Chen and Karen H. Jin (University of New Hampshire, USA)

1
Cybersecurity workforce shortage has emerged as an important concern for the nation. Although there are existing training programs for adults, cybersecurity education for K-12 students still needs significant effort. This work introduces three hands-on-oriented cybersecurity education programs: demo projects for a field trip tour, tech camp, and summer research camp. Four demo projects have been developed to motivate students. Through our outreach activities, we identified the knowledge gap between the current high school science curriculum and cybersecurity education at college. Moreover, we learned promising ways to further engage high school students to prepare for future studies in the field of cybersecurity.
Speaker
Speaker biography is not available.

AI Assisted Microcontroller Based Kits For STEM Education

Kam C Sum (AIquanta Ltd., Singapore); Kei-Hin Ng, Siu-Ting Siu, Ho-Yin Chui and Cheuk-Lut Au (AIphotonics Limited, Hong Kong); Chiu F Li (Cognitio College Kowloon, Hong Kong)

0
Artificial Intelligence (AI) is a popular topic in STEM education. Although it covers a wide range of areas, there is a strong need to start early and teach teenaged students the use of AI, inspiring them to become future scientists and engineers. In this paper, we present a recent robotic car project which is based on an Enhanced Arduino Shield Platform. The Enhanced Platform is equipped with a flexible accessory expansion structure enabling multiple devices such as motors and sensors to be added. In addition, basic AI features such as image detection and voice recognition are preinstalled on the Enhanced Platform rendering it a powerful tool for movement control of the robotic car. Students can learn basic AI features as well as coding skills through the robotic car kit. The Enhanced Platform enables us to meet the requirements of applications where were deemed too complicated or expensive for STEM education.
Speaker
Speaker biography is not available.

What Keeps Women in IT Degrees

Andreea Molnar (Swinburne University of Technology, Australia); Anne Brüggemann-Klein (Technical University of Munich, Germany)

1
Despite increased efforts to attract and support women in IT fields, their numbers remain low. Several factors contribute to this, including the retention of women once they enroll in IT degrees. In this study, we shed light on how an initiative helps women in IT fields. Through interviews with women enrolled in these fields, we found that this initiative creates a space where they can find mentorship, develop a sense of belonging, and connect with role models. Additionally, personal characteristics like confidence, self-determination, and the ability to stand up for oneself were perceived by the women as useful in this field.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-07

Works-In-Progress 07 - On-site

Conference
3:15 PM — 5:00 PM EST
Local
Mar 9 Sat, 3:15 PM — 5:00 PM EST

The A3Sat Emulator: A Catalyst in Disruptive CubeSat and Space Technology

John D Moore (Institute for Earth Observations, USA & NASA GLOBE Mission Earth, USA); Sriram Elango (Institute for Earth Observations & Harvard University, USA); Maxwell Friedman (Institute for Earth Observations, USA); Jin Kang (United States Naval Academy, USA); Christine Maceo (USNA, USA)

0
Over the past two decades, small, low-cost satellites, known as CubeSats, have rapidly progressed from research space platforms to significant mission-capable spacecraft. The capabilities of CubeSats continue to expand and are being deployed in a wide range of sophisticated scientific and commercial missions, demonstrating that CubeSats have earned a legitimate place in the New Space Ecosystem. Using CubeSats as an authentic instructional technology offers unique learning opportunities for secondary and postsecondary students to gain. hands-on experience and engage students in real-world space applications. offer a unique platform for experiential learning, empowering students to actively participate in the entire lifecycle of a satellite mission, from concept to operations to data interpretation. They provide invaluable opportunities for students to gain practical skills, foster innovation, and contribute to scientific research in space. However, the cost to build a space-ready CubeSat may exceed $100,000, prohibiting secondary student involvement in the satellite design process. The A3Sat 2.0 was designed to authentically replicate building a CubeSat for pre-college students Incorporating the A3Sat program has elevated student's satellite experience from using satellite and remote sensing databases to the world of building satellites and obtaining data, from, into the world of building small satellites and obtaining data on their own hardware.
Speaker
Speaker biography is not available.

Understanding Space Weather Through Storytelling Data Visualization

Angela Brantley (East Orange Campus High School, USA); Katherine Herbert, Vaibhav Anu, Sumi Hagiwara and Stefan Robila (Montclair State University, USA); Jason Wang (New Jersey Institute of Technology, USA)

2
Many students have difficulty understanding graphs. It may not be that they are not interested in what can be learned from a graph, rather how information is presented in a graph can be confusing. The project will focus on utilizing graphic design principles to create visual representations that effectively communicate the complex and nature of space weather. Through the incorporation of scientific data and research on solar activities, students will employ design elements such as layout, color theory, typography, and illustration to represent the various aspects of solar weather. The primary goal of this project is to create a simpler way to communicate information using visuals to understand the effect of solar activities on our planet. By merging scientific knowledge with creative design skills, students are encouraged to create informative and visually compelling materials such as infographics, posters, animations, and interactive digital media through an app. These visual representations will aim to clarify the causes, consequences, and predictive measures related to solar weather events.
Speaker
Speaker biography is not available.

Smart Design Evolution with GenAI and 3D Printing

Adrik Ray (Huber Street Elementary School, USA)

0
This paper presents a system design environment (SDE) that leverages the power of generative AI and rapid prototyping to revolutionize product design iteration. The core of the SDE lies in the GenAI Analysis Engine, an intelligent system that extracts actionable insights from diverse user feedback data using large language models (LLMs). Through a data preparation pipeline, prompt engineering module, and LLM-powered analysis, the GenAI Engine identifies recurring themes, user pain points, reported flaws and feature sentiment, informing subsequent design iterations. To bridge the gap between insights and tangible solutions, the SDE seamlessly integrates with popular CAD software and offers robust parametric design tools. This empowers designers to directly translate AI-driven insights into real-time design modifications, visualized within their familiar CAD environment. Furthermore, integrated material and functionality simulators provide valuable predictive insights, enabling informed design refinement and minimizing the need for physical prototypes. Closing the loop on user feedback, the SDE leverages a cloud-based 3D printing network to rapidly produce and deliver prototypes for real-world testing. User feedback from these prototypes is then fed back into the GenAI Engine, prompting further design refinements, and ensuring a continuous cycle of improvement. This paper demonstrates the SDE's potential to significantly shorten design cycles, reduce costs, and enhance product quality by harnessing the synergy between generative AI, rapid prototyping, and human expertise. This novel approach paves the way for a future of data-driven design, where user insights seamlessly guide product evolution, leading to superior user experiences and market success.
Speaker
Speaker biography is not available.

Maize Phenotype Classification Using GNN: Research Perspective

Avimanyou K Vatsa (Fairleigh Dickinson University, Teaneck, USA); Kevin Zhang (Bergen County Technical High School Teterboro NJ, USA); Kris G Quinche-Dutan (Fairleigh Dickinson University Teaneck NJ, USA)

0
To address the scarcity of food by 2050 for the world population and increase food productivity, we considered the maize plant as an organism. We used its phenotypes (lesions), observable and measurable traits of an organism of six different inbred lines (genetic backgrounds). The phenotypes are complex due to interaction among biochemical reactions, environmental conditions (temperature, humidity, soil quality, rainfall, etc.), and genomic expressions. The maize phenotype images are collected from field experiments over fifteen pedigrees. These image datasets are input datasets [7, 8, 15] to find a network among phenotypes and classify them. Therefore, this research uses a Deep Learning method called Graph Neural Network (GNN)), and other variants of GNN (Convolutional GNN, Graph Autoencoder (GAE), Recurrent GNN, Gated GNN, Spatial-Temporal GNN (STGNNs), etc.), and Python packages, including OpenCV Python, on the Keras framework. These are used to identify nodes and edges for finding the network among phenotypes. Also, this research provides a framework for predicting nodes, edges, network mining among maize phenotypes, and graph-related problems. Later, this outcome will help identify a group of genes responsible for this phenotype, including the yield of maize kernels.
Speaker
Speaker biography is not available.

Aviation STEM - Connecting Aviation to Education

Lyndsay Digneo (Federal Aviation Administration, USA)

0
Since 2017, the Aviation STEM Program at the Federal Aviation Administration William J. Hughes Technical Center has been successful through traditional outreach efforts by visiting schools and community organizations to provide awareness about aviation career pathways. However, program representatives wanted to make a larger impact by directly connecting outreach content to classroom lesson topics. This paper explores how the Aviation STEM Program researched curriculum standards and developed aviation-focused classroom content directly related to the standards. Additionally, this paper explains how the aviation-focused classroom content is scalable for different grade levels and how the Aviation STEM Program collaborated with the education community to multiply the impact these lessons have.
Speaker
Speaker biography is not available.

The Development of STEM Designated Programs in Business Schools: The Case of Master's Degree in Business Analytics

Bongsug Chae (Kansas State University, USA)

0
The interest in STEM-designated programs within business schools has surged, driven by the rise of big data and analytics, a data-driven corporate culture, and student visa policies. One such program is the Master of Business Analytics (MSBA), which has become increasingly popular among both domestic and international students. This research explores the current status of MSBA programs in U.S. business schools using a direct survey method. Two types of data are utilized in this research: curricular data (e.g., core courses, STEM designation) from university websites, and public institutional data (e.g., enrollment, private vs public status). This research utilizes both data visualization and statistical analysis to draw insights from the merged data. We plan to present initial findings at the conference in March.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-08

Works-In-Progress 08 - On-site

Conference
3:15 PM — 5:00 PM EST
Local
Mar 9 Sat, 3:15 PM — 5:00 PM EST

Software Defined Radios in Communications Engineering Classroom

Ashanthi S Maxworth (University of Southern Maine, USA)

1
This work-in-progress paper describes the implementation of software-defined radio-based experiments for ELE 483: Communications Engineering at the University of Southern Maine. Starting the spring of 2024 this course is required for both electrical and mechanical engineering undergraduates. This new integration of disciplines poses new challenges to the instructor in terms of demonstrating the applicability of the course, in mechanical engineering in addition to electrical engineering. Given that the course contains heavy mathematics, the best way to show the applicability of communications engineering in day-to-day applications such as radio broadcasting, and aircraft navigation. The fastest and the most cost-effective way to design these hands-on experiments is by using software-defined radios or SDRs. The SDRs are being used in hobby applications to high-end research applications. This work-in-progress paper describes the content of the course and the SDR-based experiments that will be integrated into the course.
Speaker
Speaker biography is not available.

Understanding Cyberbullying Patterns Utilizing Word Clouds

Brionna Nunn and Soo-Yeon Ji (Bowie State University, USA)

0
Social media provides a convenient way to communicate among people by sending and receiving content. However, children and teenagers may suffer from emotional, mental, and behavioral problems due to its negative influences, such as unpleasant, damaging, deceiving, or malicious content towards an individual, referred to as cyberbullying. Thus, detecting suspicious conversations in cyberspace is essential for users to increase their awareness. This research aims to identify important features from cyberspace texts to generate a predictive model. As an initial step of the study, we focus on understanding the patterns between cyberbullying and non-cyberbullying utilizing word clouds. In detail, after pre-processing steps such as text cleaning, stopword removal, and lemmatization, we compared two measurements, term frequency-inverse document frequency (TF-IDF) and Bag-Of-Words (BoW), to determine the importance of words in identifying cyberbullying by selecting the most frequently used words. Word clouds are then created to present the selected words visually. The determined words could be used as essential indicators to determine cyberbullying conversation. We also compared emotion-related words between cyberbullying and non-cyberbullying texts. This study can provide a deeper understanding of language patterns related to cyberbullying conversations.
Speaker
Speaker biography is not available.

An Acceptance Index for Mathematical Software tools to Motivate Women pursuing STEM Studies

Olga Lopez and Marcia Lechuga López (Instituto Tecnológico y de Estudios Superiores de Monterrey, Mexico)

0
One critical aspect that has gained attention in recent research is the role of mathematical tools and technological applications in shaping women's decisions to pursue STEM careers. Understanding the factors that influence women's acceptance of mathematical tools and their subsequent impact on career choices is essential for fostering gender diversity in STEM. Dozens of studies reveal that women avoid STEM careers because they perceive them as too demanding or involving excessive mathematics. Over the last decades, mathematical software and applications have been developed, facilitating the use of mathematics at the undergraduate level. In this study, we introduce a novel approach-the Mathematical Acceptance Index. This index is designed to evaluate and assess the acceptance and use of mathematical software and tools among young women. It measures the impact on the interest in using mathematical software and applications to solve real-life problems requiring only basic knowledge in mathematics and statistical concepts. Our index was built based on research on acceptance models of new technologies. The study involved the evaluation by a group of female students, focusing on the ease of use and perception of usefulness of three mathematical software-Mathematica, Geogebra, and SPSS. These tools were applied to solve simple, yet real problems involving basic knowledge of algebra, calculus, and statistics. The study aimed to measure the perception of female students regarding their intention to continue in a STEM career. Initially, a survey assessed a structural equation model on the same student sample to gauge their interest in pursuing a STEM career. We present our findings on the structural equation model first, followed by our methodology and the results concerning the Mathematical Acceptance Index.
Speaker
Speaker biography is not available.

Analyzing the Health of Lithium-ion Batteries through Heat Distribution and Thermal Modeling

Rohit Karthickeyan (John P Stevens High School, USA); Sushanth Balaraman (Edison High School, USA)

0
For decades, the primary focus in battery health assessment has been on metrics such as voltage levels and current flow. However, ThermoBatt shifts the lens towards the thermal attributes, a domain less explored but equally vital. ThermoBatt encompasses two innovative models: the first, a machine learning algorithm, predicts the State of Health (SOH) and Remaining Useful Life (RUL) of batteries by analyzing factors such as ambient temperature and usage cycles. The second, a real-time temperature distribution model, utilizes temperature data within charge/discharge cycles to simulate thermal behavior. This approach necessitates several assumptions, underscoring the pioneering nature of our exploration. ThermoBatt aims to deepen our understanding of how heat generation and distribution influence battery health and longevity. By bridging this knowledge gap, our work illuminates the interconnectedness of thermodynamics with battery efficiency and endurance, paving the way for advancements in battery technology and sustainable energy solutions."
Speaker
Speaker biography is not available.

Engaging Middle School Students with Disease Ecology in Place-based STEM Outreach

Tugba Boz (Purdue University, USA); Nora Smith (Montana State University, USA); Rebekah Hammack (Purdue University, USA); Hilarie Davis (Technology for Learning Consortium, USA); Jamie Cornish (Montana State University, USA)

0
In this paper, we report on a place-based workshop on disease ecology as part of a nationally-funded project. Our approach to disease ecology is interdisciplinary encompassing elements of ecology, microbiology, medicine, genetics, immunology, entomology, geography, nutrition, and epidemiology. In this paper, we describe how middle school students, participating in a summer camp at a university in the northwestern U.S., engaged with a five-day workshop on disease ecology. We inquired into how the students' engagement with and understanding of disease ecology were enhanced through place-based activities, and which aspects of these activities were meaningful for their learning about disease ecology. We had two sections and a total of 24 students in the workshop. Our participating students consisted of 13 white students, 11 Native American students, and one Hispanic student. Our study utilized two primary sources of data: student interviews and student-generated artifacts. Our analysis of these interviews and artifacts employed a multi-method approach to ensure a comprehensive understanding of the data. Our preliminary findings show that the students found the camp activities engaging and that these activities increased their interest in science and health. The place-based aspect of the activities increased their understanding of science in their local communities and contexts.
Speaker
Speaker biography is not available.

Improving Computational Thinking Competencies in STEM Higher Education

Nasrin Dehbozorgi (Kennesaw State University, USA); Mehdi Roopaei (University of Wisconsin - Platteville, USA)

0
Computational thinking (CT) is a key competency with a significant impact on students' academic performance, particularly in STEM fields. It empowers students to enhance problem-solving skills by decomposing problems, utilizing abstraction and pattern recognition, and employing algorithmic thinking to design solutions and build models. This is particularly important in STEM disciplines where logical reasoning is essential for addressing complex real-world challenges in academic and industrial settings. Given the increasing demand for professionals equipped with strong algorithmic thinking and problem-solving abilities in Industry 5, educational institutions are focusing on enhancing students' CT and problem-solving skills. This study presents an initiative conducted over the past two years at our institute to teach CT in a gateway course to students with any background in STEM fields. The approach involved designing specific learning modules on Abstraction, Decomposition, Pattern Recognition, and Algorithmic Thinking and integrating them into the LMS. After studying these learning modules, the students were exposed to specific assignments that required application of related CT skills. Pre and post surveys were employed by using standard CT tests to measure the impact of the intervention on students' CT levels. The results indicated an improvement in students' perceptions of their mastery in CT. Academic course grades also showed an improvement, with increased A and B scores and reduced C and D grades post-intervention. This two-year study on improving CT skills has yielded promising results. Moving forward, the research aims to enhance the existing modules further and distribute them to a broader range of introductory-level STEM courses across the university. This future direction aligns with the goal of expanding the impact of CT education and integrating it more widely into STEM curricula.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-09

Works-In-Progress 09 - On-site

Conference
3:15 PM — 5:00 PM EST
Local
Mar 9 Sat, 3:15 PM — 5:00 PM EST

Creating a Cross Curricular Resource for Solar Weather History and Its Impact on Daily Life

Margaret Mary S. Menichella (Passaic Public Schools, USA); Stefan Robila and Katherine Herbert (Montclair State University, USA); Thomas J Marlowe (Seton Hall University, USA)

2
As our society becomes more interdependent on electronic communication, we become more vulnerable to the impact of Solar Weather on our everyday lives. When researching this topic, information is not readily available for a targeted audience of the average middle school student. This paper outlines the content and process for creation of an interactive web resource, compatible with the most common accommodation tools, that will highlight how common solar weather events are and how rarely they have an impact on our lives due to the work that is already being done to protect us. This resource will be cross curricular and highlight both the scientific impact such events have had but also the social and societal impacts of the phenomena for at least 15 of the most referenced events.
Speaker
Speaker biography is not available.

K-12 Teachers' Perceptions of Computer Scientists: Insights from an Equity-Oriented Professional Learning Community

Minsun Shin, Sumi Hagiwara, Katherine Herbert, Vaibhav Anu and Rebecca Goldstein (Montclair State University, USA)

2
Guided by visual and content analysis, this work-in-progress paper explored the perceptions of kindergarten to grade 12 teachers regarding computer scientists and how participation in professional learning communities has influenced their perspective on the field of computer science. Sixteen educators participated in an equity-oriented professional learning community hosted at a major university in New Jersey. An interdisciplinary team of university faculty in computer science and teacher education collaborated to share their relative areas of expertise and develop the equity-oriented and inclusive 2-year grant program. Qualitative content analysis of a draw-a-computer-scientist task revealed that participants depicted computer scientists actively engaged in computer usage, binary coding, and programming. The results revealed that participants overwhelmingly held a belief that anyone, regardless of gender, race, or age, has the potential to become a computer scientist. Given the commitment among participants to promote gender equity and inclusivity, this study documents and argues that well-structured professional learning community programs are crucial for teachers to reflect on their beliefs, broaden their perspectives on computer science, promote equity and inclusivity, and empower them, especially female teachers and underrepresented minorities, to become role models for their students.
Speaker
Speaker biography is not available.

The APP Method: Self-Regulation Strategies Giving POWER to Computer Science Students

Corina S Drozdowski (Glen Ridge High School & Montclair State University, USA); Solomon Emeghara (Montclair State University & Paterson Public School, USA); Thomas J Marlowe (Seton Hall University, USA); Katherine Herbert, Vaibhav Anu, Sumi Hagiwara and Stefan Robila (Montclair State University, USA)

2
Research shows that improving students' understanding of Computational Thinking (CT) in unplugged, science-based courses benefits their understanding of algorithmic thinking. This paper describes use of the APP approach, a UDL-based guideline of self-regulation for problem formulation and problem solving, and then presents a program of lessons designed to improve a student's CT skills through improvement of executive functioning and higher-level cognitive skills, using the APP framework to explore the field of solar weather, an underrepresented but relevant topic for 6-12 grade students. The approach has strong resonances with the POWER approach to expository writing, the scientific method, and software engineering.
Speaker
Speaker biography is not available.

The Development and Implementation of a Cost-effective Educational Robotic Arm using ROS-MoveIT

Matthew F Eaton (High Point University & Kennesaw State University, USA); Muhammad Hassan Tanveer (Kennesaw State University, USA)

1
Modern robotic arms have reached remarkable levels of sophistication and capability, offering unparalleled performance in various applications. However, their widespread adoption in STEM-based classrooms and engineering labs is often hindered by exorbitant costs, making them inaccessible in significant quantities for many educational institutions and students. This paper addresses this challenge by presenting the design and development of an affordable open-source robotic arm, aiming to match the performance of high-end counterparts while significantly reducing the financial barrier to entry and providing educators with exciting content for mechatronics based lab activities. The proposed robotic arm not only opens doors for students to engage in robotics education but also provides an opportunity for customization. Its cost-effectiveness enables broader participation of students in practical robotics projects and labs, fostering a deeper understanding of the entire engineering design process, including motor choice, computational platform choice, and material selection. From assembly to computer-aided design (CAD), students gain hands-on experience, enhancing their skills and knowledge in robotics. Moreover, the affordability of this robotic arm promotes inclusivity within educational settings. With a lower financial burden, institutions can facilitate the involvement of a larger number of students, fostering a collaborative and innovative learning environment. Additionally, this robotic arm project seamlessly integrates with modern robotics frameworks such as Arduino, Raspberry Pi, Robot Operating System (ROS), and MoveIt, offering students the chance to explore and understand the intricacies of real-world robotics applications.
Speaker
Speaker biography is not available.

Design and Implementation of a STEAM Robotics Lesson on the Spotted Lanternfly: Engineering a Computer Science Solution

Esther Douglass (Hillsdale Schools, USA); Katherine Herbert, Vaibhav Anu and Sumi Hagiwara (Montclair State University, USA); Thomas J Marlowe (Seton Hall University, USA); Alaina Cannella (Hillsdale Schools, USA); Stefan Robila (Montclair State University, USA)

2
Introducing middle school students to STEAM projects can foster an interest in computer science and engineering solutions. These projects provide hands-on experiences that may encourage students to explore careers in science, technology, engineering, art, and mathematics. Engaging students in real-world problems can help them develop critical thinking skills and generate innovative ideas. STEAM projects can also spark students' curiosity and creativity. This paper will explore developing and implementing a STEAM lesson to introduce students to the engineering and design process and computational thinking. Moreover, this project also addresses a major environmental concern currently in New Jersey, specifically the destruction of the environment by the spotted lanternfly. Therefore, students are also applying skills in civil duty and environmental awareness as a part of their design and plan to eradicate this invasive species.
Speaker
Speaker biography is not available.

Optimizing Large Language Models for Auto-Generation of Programming Quizzes

Yulia Kumar, Anjana Manikandan, Jenny Li and Patricia Morreale (Kean University, USA)

1
The study presents a comprehensive analysis of Large Language Models (LLMs) like ChatGPT in the creation of educational content for Java programming courses, specifically Object-Oriented Programming (CS1) and Data Structures (CS2). Throughout a semester, these models were utilized to develop and refine programming quizzes for the mentioned courses. The research is centered around three principal questions: First, evaluating how accurate the assessments are created by LLMs; second, determining the advantages and disadvantages of integrating LLMs in CS1 and CS2 education from the educators' perspective; and third, identifying best practices and best prompt engineering techniques to improve the quality of quizzes and tests generated by LLMs. The study meticulously evaluates the quizzes produced by LLMs for their consistency with Java programming's core principles, their alignment with specific learning objectives of CS1 and CS2 courses, and their effectiveness in enhancing student engagement and understanding. A comparative analysis is conducted between human-generated and AI-generated content, focusing on assessing the quality, complexity, and pedagogical impact, thereby offering insights into the capabilities and limitations of LLMs in academic assessment generation for computer science education.
Speaker
Speaker biography is not available.

Seeding Deception: Investigating the use of GANs in Minimizing Backdoor Poisoning Attack Ratios

Cynthia C Zhang, Akira Yoshiyama and Celena Gu (University of Waterloo, Canada)

0
As machine learning becomes widely adopted in industry for mission-critical systems, significant research is currently dedicated to the concern of cybersecurity attacks on open-source data. As we see an increasing dependency on open-source databases in model training for institutional use, there is a similarly increasing number of ways for attackers to exploit the Internet - the training ground for such machine learning models. Our investigation centers around using a generative adversarial network (GAN) to minimize the number of poisoned data elements in order to trigger a model into generating inaccurate results. In this paper, we focus on one type of adversarial attack – the backdoor attack, where the attacker provides poisoned data to the victim to train the model with, and then activates the attack by showing a specific small trigger pattern at test time (e.g. a small patch of pixels on an image). The poisoned data is mixed in with benign data, according to a certain ratio, referred to as the poisoning ratio. In particular, we aim to use the GAN in order to optimize the trigger pattern (i.e. pixel mask) added to corrupted data samples and minimize the poisoning ratio (i.e. minimize the number of corrupted samples in a model's training dataset) for a binary image classification convolutional neural network (CNN). The perspective of this research study focuses around the implications of targeted cybersecurity attacks on open-source datasets. It is therefore of utmost importance that researchers understand the mechanics of various machine learning attacks, and push the envelope on state-of-the-art attacks, such that researchers and engineers can proactively create defenses. This is a principle motivation behind white-hat hacking and behind this paper.
Speaker
Speaker biography is not available.

Enter Zoom
Session WIP-10

Works-In-Progress 10 - On-site

Conference
3:15 PM — 5:00 PM EST
Local
Mar 9 Sat, 3:15 PM — 5:00 PM EST

A Cyber-Physical Systems Approach to Teaching Solar Weather Topics in Middle School

Solomon Emeghara (Montclair State University & Paterson Public School, USA); Corina S Drozdowski (Glen Ridge High School & Montclair State University, USA); Stefan Robila, Sumi Hagiwara and Katherine Herbert (Montclair State University, USA); Thomas J Marlowe (Seton Hall University, USA); Vaibhav Anu (Montclair State University, USA)

2
While K-12 student's exposure to computational concepts continues to grow, many topics continue to be disconnected from real life problems. Computer science concepts are either generalized or the practical examples used are significantly different than real life scenarios. This is particularly the case in science where complex phenomena need to be tackled with integrated approaches that include sensing, data collection and processing. Often found in engineering, cyber-physical systems (CPS) are systems designed to seamlessly integrate physical components and computational algorithms. In this paper we describe the use of CPS to bridge science and computation in designing educational activities for middle school students. The classroom activities were designed in two steps: the first using traditional approaches that focus on modelling, summarization and investigation, while the second phase focuses on integrated system development. Using off the shelf components such as Arduino, the students are asked to develop a computational platform that combines sensing (such as temperature or magnetic field) with data collection and processing. Solar weather is used as the science case due to its significant impact on the Earth. Exposing middle school students to know how to detect the changes in environmental temperature which could help predict the occurrence of these activities of the sun is the crux of this project.
Speaker
Speaker biography is not available.

Exploiting RF Signature in 5G Simultaneous Localization and Mapping

Sean Kim and Bumsuk "Brian" Choi (Johns Hopkins University Applied Physics Laboratory, USA)

0
The evolution of modern cellular communications has been marked by a series of technology generations. With fifth-generation (5G) being deployed since early 2019, the 5G-based network has been extensively used by commercial telecommunication service providers for its ability to employ a large number of small cells, allowing more frequency re-use. Moreover, the mmWave frequency band along with directional beamforming has been known to create a challenge in detecting mobile subscriber signals and target users due to lower powers and poor propagation (e.g. high path loss, sensitivity to blockage, dynamic due to user mobility, etc.), especially for GPS-denied environments such as urban areas and building interiors. To accommodate these challenges, there have been increasing research efforts to seek reliable geolocation capabilities to ensure seamless network performance. In this work, an RF signature-aided Simultaneous Localization and Mapping (SLAM) framework is proposed to provide improved geolocation performance for 5G systems. Specifically, we first investigate the currently existing SLAM framework for 5G systems. Then, we introduce a novel radio map based on an RF signature generated using a 3-D ray tracer and integrate RSSI positioning based on the radio map with SLAM framework. Our simulation results demonstrate the power of the proposed approach in enhancing the overall localization performance.
Speaker
Speaker biography is not available.

Summer Grilling and High Flying: Modeling the Heat Transfer of Steak and Aerospace Vehicles

Sri Suhas N Chokkaku (Johns Hopkins University Applied Physics Laboratory (JHUAPL), USA); Samuel Chen (Johns Hopkins University Applied Physics Laboratory, USA)

0
Aerospace vehicles traveling at extremely high speeds experience intense temperatures and pressures from the air around it. At these conditions, aerospace vehicles are exposed to a variety of different heat transfer mechanisms that can heat up the vehicle's exterior and potentially alter its chemical composition. To provide an analogous model of the heat transfer of aerospace vehicles, this paper analyzes the heat transfer problem in a more common household scenario: cooking the perfect steak. To answer this question, the steak's heat transfer is modeled using the 1-Dimensional Heat Equation with different boundary conditions representing different cooking methods, such as using a convection oven or a cast-iron pan. Each method utilizes either Dirichlet boundary conditions which directly impose a temperature to cook the steak, or utilizes Neumann boundary conditions which impose a temperature gradient to initialize the cooking process. From a physical perspective, this study emphasizes the need to take into account for the different "rarities" of steak as different cooking conditions may be needed to achieve those "rarities". Similarly, there are a plethora of conditions which could affect the heat transfer of aerospace vehicles, so this study is meant to demonstrate a numerical framework to predict heating to a vehicle's exterior.
Speaker
Speaker biography is not available.

VOYCE: Transforming Public Speaking Apprehension into Confident Advocacy for Community Advancement

Archishma Marrapu ( & STEMifyGirls, USA)

0
This research investigates the transformative impact of the VOYCE app on individuals grappling with public speaking apprehension, seeking to evolve their fears into tools for confident advocacy. As an innovative platform, VOYCE addresses fear management through personalized feedback, diverse practice modules, and a supportive community. The study explores how users, empowered by the app, transcend their speaking anxieties, ultimately becoming articulate advocates for community advancement. By analyzing the app's role in fostering confidence and communication skills, this research contributes to the broader discourse on the intersection of technology, personal development, and societal progress. Through VOYCE, individuals not only conquer their fears but also harness the power of their voices to make a meaningful impact on their communities.
Speaker
Speaker biography is not available.

Advancing Skill-Based Pathways to Good Jobs in Artificial Intelligence at Community Colleges

Yasmin Saaid (San Antonio College, USA); Heena Rathore (Texas State University, USA); Henry Griffith (San Antonio College, USA)

0
This paper provides provides recommendations for developing new skill-based employment pathways in AI at community colleges. These recommendations are motivated by a summary assessment of multiple influencing factors, including the composition of the current AI workforce, employers' perception regarding skill-based hiring, and implementations of existing pathways at two-year institutions. Our analysis indicates that pathway development should be expedited in order to capture favorable trends in labor demand and hiring protocols. Leveraging curriculum developed by industry leaders, and ensuring the integration of skill validation mechanisms, including badges and industry-recognized credentials, are also suggested as a means of addressing existing barriers for individuals without four-year degrees.
Speaker
Speaker biography is not available.

Impact of Quantum Mechanics-Based Workshops on Developing High School Students' Interest and Intuition in Quantum Information Science

Padmanabh Kaushik (Lafayette College, USA); Nam Vu (Lafayette College & Yale University, USA); Crystal Yeung and Nicholas Sorak (Lafayette College, USA); Pouya Khazaei (The University of Michigan, USA); Delmar Azevedo Cabral, Brandon C. Allen and Victor S. Batista (Yale University, USA); Heidi Hendrickson (Lafayette College, USA)

0
Our research aims to analyze the impact of quantum mechanics-themed games on the development of high school students' intuition and interest in topics related to quantum phenomena, such as quantum superposition and quantum measurement. In the United States, quantum concepts are introduced at a fragmented level in high schools, and efforts to implement a quantum curriculum at the secondary level have been limited. This presentation describes the development of a workshop, and the associated research study, scheduled for implementation in the late winter and spring of 2024. The workshop utilizes an interactive game-play method to introduce students to concepts in quantum computing. We anticipate that this workshop-based game-play pedagogical approach will enhance students' ability to connect intuitive and non-intuitive concepts in QISE (Quantum Information Science and Engineering) to their experience of playing the game. We hypothesize that enabling high school students to interact with quantum mechanics concepts during their secondary education will foster sustained interest and potentially encourage them to pursue further studies in the field of QISE.
Speaker
Speaker biography is not available.

Accessible Control of a Robotic Arm for People with Physical Mobility Limitations

Mark Amidon, Kyle Patterson and Girma Tewolde (Kettering University, USA)

0
Conventional electronic gaming controllers lack the necessary features that make them accessible for people with physical mobility limitations. Children and adults who have difficulties to use their two hands or who may have limited motor skills could find it challenging to play video games or operate remotely controlled devices using ordinary controllers that require the players to use multiple fingers from two hands. The objective of this project is to create an accessible solution to control a robotic arm utilizing two alternative control methods – using an Xbox adaptive controller and a head tracking system. This is a collaborative project between Kettering University and the Genesee Intermediate School District (GISD). The GISD Special Education Services Program offers educational programs for students who require special accommodations, due to physical or other impairments. The proposed project demonstrates effective prototype for an accessible robotic control solution for students who may not be able to use traditional gaming control device.
Speaker
Speaker biography is not available.

Digital Twins of Rare Metals Production Based on Neural Network Technologies for Complex Process Automation in Non-Ferrous Metallurgy

Gulnara Abitova (State University of New York at Binghamton, USA & Astana IT University, Kazakhstan); Vladimir Nikulin (Binghamton University (SUNY), USA)

0
The search for new methods and approaches for optimizing modes, effective process control systems are relevant tasks. One of the directions of digital transformation and industry modernization is a digital twins design of production facilities, which allow real work simulating in a different modes and scenarios. Thanks to the "virtual machine", production facilities can optimally adjust and predict the equipment operation, real time control processes, produce an appropriate quality product. A promising area of digital twins' application is a complex technological processes modeling in the production of rare metals automation. This is due to the multicomponent nature of raw materials, a transient processes multitude, a physical-chemical reactions complexity, decision-making issuers. There is a gradual depletion of ores components, causing significant regimes deviations. The study is based on the processes mathematical modeling, development of a processes digital copies, predictive data analytics, real-time decision making using NN. The digital twins based on NN will ensure the efficiency and reliability of control systems, to predict an equipment failure. The industry digital twins will reduce a production cost, improve a product quality.
Speaker
Speaker biography is not available.

Enter Zoom