Game2 Learn Research Lab: go.ncsu.edu/g2l
Serious Games: Games are frequently the medium of choice for training and education and promoting awareness and behavior change because of pervasive technologies and games’ inherent motivational qualities and their increasing capability to simulate real situations. REUs will design, develop, and evaluate their own serious games for socially relevant purposes. To prepare, they will read relevant literature and improve existing lab games (current games include BOTS and Resource Rush). They will also help conduct studies on existing serious games in summer camps. REUs will develop an independent research question, such as “Can we improve math learning by situating it in a game context?” or “How can we create logs to help understand player behavior in this particular game?” Their research has led to research contributions in human-computer interaction (CHI), AI in Interactive Digital Entertainment (AIIDE), Foundations of Digital Games (FDG), etc., as the referenced works above.
Data-Driven Tutoring: Computer learning environments that mimic aspects of human tutors, called Intelligent Tutoring Systems (ITSs), have been shown to be highly effective in improving students' learning in real classrooms (Vanlehn, 2005). REU students will assist in designing, developing, and evaluating effective intelligent educational environments whose intelligence is derived directly from log data. In particular, we are refining tutors for three important undergraduate stem domains: logic, probability, and introductory programming, contributing to Barnes’s extensive work in this area (e.g., Mostafavi, 2017; Price, 2016; Cody, 2022; Shabrina, 2023). This work is socially relevant as it improves education for STEM-related fields. REUs will perform data analytics on log data to create hints, supports, and student learning and affect models from data. REUs on these projects will work with Barnes as a world leader in data-driven tutoring with research contributions in artificial intelligence in education (AIED), educational data mining (EDM), computer science education (SIGCSE, ICER, ITiCSE, Koli, WiPSCE).
Computing Education Research Engaging All Learners Lab: website
Culturally Relevant Simulations and Games. CS Education lessons are conducted in magnet schools, rural schools, and schools with limited resources. Applying a one-size-fits-all approach to these diverse ecosystems leads to a disconnect between the stakeholders, researchers, and lesson content. For example, parents near the Appalachian Mountain region dislike 'computer science' because it implies their children will move to Silicon Valley; these parents prefer ‘cybersecurity,’ which is taught in the surrounding colleges and leads to good careers near home. While designing and developing interactive simulations for K-12 classrooms, REUs will not only make learning more engaging but also enhance cross-cultural competence. REUs will use stakeholder feedback from prior lesson implementations to develop a range of simulations and games that represent various cultures, traditions, and historical contexts (e.g., medical delivery drones in Rwanda). REUs will develop the technical aspects of the simulations and games, including coding, graphics, and user interfaces, as well as conduct user testing sessions from which they can analyze and later present the data. REUs will pilot their creations in a low-stakes informal learning environment, such as a summer camp that will provide quick feedback for design iterations. REUs will analyze their results on learners' attitudes and cognitive changes from surveys and interviews and present them at the final presentations. This research contributes both to rapid software engineering design cycle practices as well as to findings on how intentional awareness of culture can enhance student attitudes and learning in CS.
Analytics and Machine Learning for Education and Healthcare Lab: Website
Dr. Chi’s lab is focused on designing and applying more advanced machine learning and data mining algorithms that enable the behavior of a computer to be learned from examples or experience rather than dictated through rules written by hand. We will teach students how to use advanced machine learning algorithms such as deep learning models, reinforcement learning, and deep reinforcement learning to analyze interactive data, such as mining user log files for predictions, healthcare data, and food bank data, finding patterns in sequential data, or building adaptive interactive systems.
Generalizing Data-Driven Technologies to Improve Individualized STEM Instruction by Intelligent Tutors. In the end, this work will provide students with adaptive, individualized support at multiple granularities through the use of a modular framework of educational data mining methods that are implemented, iteratively refined, and empirically validated for learning impact and robustness across systems. A general hierarchical data-driven framework will be used by REU participants to develop a general platform that addresses the conceptually and practically complex task of constructing adaptive learning support. REUs will demonstrate the platform's effectiveness across three STEM domains, including logic, probability, and programming, where traditional ITs are extremely challenging.
Early Diagnosis and Prevention of Sepsis. The project data set includes all Mayo Clinic - Rochester hospital discharged patients over a 5-year period, as well as all Christiana Care hospital discharged patients over a 2-year period. The Mayo Clinic sample from 2010-2011 includes 47 million observations from 79K patients during 571K hospital days, while the CCHS sample includes approximately 120,000 visits for 115,000 patients. As part of the REU program, students will develop and apply novel Deep Learning and Deep Reinforcement Learning frameworks to model the sepsis spectrum and to induce effective interventions to prevent sepsis. In addition to integrating Health Records and clinical expertise, this project will provide an evidence-based framework to diagnose patients within the sepsis spectrum and develop and validate clinical interventions. REU students will analyze real-world data, develop student models, and develop decision-making policies based on real-world data.
Human-Centric Software Engineering Lab: Website
Human-centric software engineering lab focuses on the human aspects of software engineering by studying and modeling programmer behavior and then designing and developing mixed-initiative programmer-computer systems. The lab's pursuit of excellence is underscored by its multidisciplinary approach, amalgamating the domains of Human-Computer Interaction, Software Engineering, and Artificial Intelligence. By synergizing these diverse fields of expertise, the lab pioneers the development of novel strategies, innovative theories, immersive visualizations, and tangible prototypes, all tailored to cater to the unique needs of programmers.
Ethical ChatGPT: Enhancing Programming Proficiency seeks to address the critical need for a software solution that enhances programming education by guiding students through a comprehensive learning experience. The tool will aim to bridge the gap between utilizing code generated by LLMs like ChatGPT ethically by making students truly understand the essential concepts that underpin proficient programming. The tool should focus on reinforcing the understanding of the time and space complexity of programs while also encouraging the exploration of diverse data structure implementations for a given problem.
Fostering Gender-Inclusive Pair Programming Inclusive PP Tool: The project aims to develop powerful software to address gender-based dynamics in pair programming. This tool will transform pair programming by analyzing communication styles, leadership roles, interruptions, and partner preferences. The goal is to build understanding and empathy between pairs, enhancing collaboration, code quality, and productivity, regardless of gender composition. This project's main goal is to create a facilitator agent for better interactions between same and mixed-gender pairs. The system should carefully monitor how individuals, both in same-gender and mixed-gender pairs, collaborate during pair programming, roles of individuals, observing communication patterns, leadership dynamics, and interruptions. The system should also allow real-time data capture without disrupting the programming process. Additionally, an analytics engine processes the gathered data to create visualizations and insights, which is presented in a user-friendly format to promote better self-awareness and understanding of their collaboration patterns. REU students will assist in the design, prototyping, and development of the ethical ChatGPT and gender-inclusive Pair Programming tools, and in conducting studies and perform both quantitative and qualitative analytics on study and tool datasets.
Center for GeoSpatial Analytics: website
Dr. Pala's research group investigates ways to analyze, visualize, and provide education on energy, food, water, and transportation systems and their complex relationship with our everyday lives. The group has a special focus on the systems’ interrelations and the two-way relationship of the members of the public with each system.
Food, Energy, Water (FEW): These projects apply advanced geospatial analysis techniques coupled with cutting-edge computing approaches. They analyze, simulate, and visualize regional and national energy, water, and transportation data to help guide decisions regarding system planning, local and regional policy-making, and environmental problems. Dr. Pala’s lab also focuses on the effects of transportation infrastructure changes on future urban expansion, along with the equity issues that might arise due to the implemented changes.
Develop geospatial analysis tools: These tools help create future urban expansion projections and identify areas of concern regarding equity and access due to existing/future transportation infrastructure. Past projects include a study that identified areas where people affected by specific transportation projects are located to implement focused public outreach and the development of spatial decision support systems for critical infrastructure outage management when cascading outages affect multiple interrelating infrastructures. REU students will assist in the development of interactive dashboards that reference data from the National Renewable Research Lab’s Regional Energy Deployment System (ReEDS) and FEWSION research group’s county-scale commodity flow data that allows users to navigate and explore the content while following specific scenarios. REUs will also conduct studies and analyze data from surveys, screen captures, and voice records to compare the effects of scenario-based interactive learning on the participants’ understanding of the problem space and possible solutions, behavior, and interest in STEM fields.
Help through INTelligent Support (HINTS) Lab: go.ncsu.edu/hintslab
The HINTS Lab research goal is to re-imagine educational programming environments as adaptive, data-driven systems that support students automatically as they pursue learning goals that are meaningful to them. His work has focused on the domain of computing education, where he has developed techniques for automatically generating programming hints and feedback for students in real-time by leveraging student data. His HINTS lab focuses on supporting students working in creative, open-ended, and block-based learning contexts, leading to novel data-driven programming support, including adaptive examples, subgoal feedback, and models to predict student outcomes.
Understanding Novice Programmers’ Help-Seeking Behaviors in the Age of ChatGPT: Programmers make frequent use of external resources to learn new languages, frameworks, and paradigms. Previously, these external resources can include things such as documentation and Q&A forums such as StackOverflow. However, with the advent of modern AI code generation tools such as ChatGPT, it is unclear how all of these resources should be integrated into computing curricula. Proficient use of these external resources for just-in-time learning is a critical skill for programmers, yet the methods in which novices use and learn these tools is understudied. REU students will have the opportunity to analyze data related to how novices use these resources, ranging from programming log and interview data on novices’ reactions to using web-search while programming to large-scale survey data collecting students’ beliefs and perceptions on external resources. REU students will also use this data to build statistical models in order to better understand what factors influence novice behavior and use these results to inform prototype tools and curricula interventions to better teach these help-seeking behaviors.
Improving Novices' Motivation to Learn Programming: Research in computer science education and student psychology has demonstrated that motivation is a key component of student success; however, studies often focus on understanding motivation through its effects (i.e., performance and retention) rather than through the factors that contribute to increased or decreased motivation. As a result, while we understand the importance of motivation, we do not understand how to better motivate students. REUs will use preliminary data from a study evaluating the importance of various factors on student motivation to construct a model of how those factors affect motivation, but also how those factors affect each other. REU students will propose and design an intervention to increase motivation based on the developed model. This project will allow us to develop a CS-specific model of factors affecting motivation and contribute to research in educational psychology and CS education.
Generative Intelligent Computing Lab: Website
Dr. Xu’s projects focus on generative AI technologies and their applications in K-12 education.
Generative AI for Creative Writing in K-12 Classrooms: This research project invites REU students to explore the potential of ChatGPT in facilitating creative writing exercises in K-12 educational settings. The goal is to fine-tune the generative AI model to create age-appropriate and genre-specific creative writing samples, such as poems, short stories, and essays. REU students will contribute to designing, building, and studying generative AI tools, conducting user studies with the tools, and generating creative writing pieces using ChatGPT. REU students will leave with a comprehensive understanding of how generative AI can be employed in the field of creative writing, specifically tailored for K-12 educational settings.
Crafting Generative AI Tools for K-12 Educational Ecosystem: This project offers REU students the opportunity to utilize ChatGPT and other generative AI technologies like DALL-E to develop interactive educational tools. These AI-powered platforms aim to enhance learning experiences across various K-12 subjects, including math, science, history, and language arts. REU students will contribute to tool design, development, and user testing with K-12 students and teachers to assess the effectiveness and safety of the generated content. REU students will gain hands-on experience in creating generative AI tools that can significantly impact K -12 education, focusing on personalized and adaptive learning experiences.