Data-driven Intelligent Tutoring Systems

This project will augment three existing e-learning interactive environments; more specifically, it will add data-driven techniques for the automatic generation of next-step hints and for the automatic selection of learning activities.

Deep Thought is an intelligent tutor for the practice of solving deductive logic proof problems in graphical representation displaying the proofs as logical premises, with buttons for logical rules that can be applied to selected premises and derived expressions, and a logical conclusion as the goal of the problem. We have been incrementally augmenting the Deep Thought logic tutor with data-driven methods for formative feedback and hint generation, problem selection, and worked examples to improve student learning of logic proof solving and reduce tutor dropout. Our long-term goal is an intelligent tutor for logic-proof construction that is fully data-driven and can adapt to students' learning logic with varying curricular requirements without the need for further expert input.

This project will develop hierarchical data-driven, interpretable, and robust models that optimize human learning. Moreover, it will investigate whether integrating hierarchical data-driven agent decision-making with user-initiated decisions can help students learn to make better decisions for their learning.

Educational Data Mining

The ModSoc project is a collaborative research project focused on the identification of useful behavioral patterns in online educational tools. The project is joint work between ourselves, and researchers at the Teacher's College at Columbia University, The Educational Testing Service, and Arizona State University. As part of this work we are collecting repositories of data from blended courses offered at NCSU and UC. Berkeley as well as MOOCs offered through the Teacher's College. We have already completed the analysis of meaningful student communities in this data and shown that students form at-will communities that can reflect their own subsequent performance. We are also examining the role of peer tutoring in this data as well as the paths students take through online materials.

The IUSE project is focused on the extraction of pedagogical and conceptual hints from existing tutoring technologies. This is joint work by Dr. Min Chi, and Dr. Tiffany Barnes. Dr. Chi's work has previously been focused on the extraction of pedagogical guidance from student data and she is continuing that work here to extract pedagogical guidance from the existing logic tutoring data. Dr. Barnes' work is focused on the extraction of conceptual information from new domains such as probability.

Computer Science Education

SnapClass is a research and learning management platform for the block-based programming language, Snap!. Students can program and submit their assignments in SnapClass and teachers can assign, create rubrics, and grade within the environment. Through participatory design with K12 instructors, SnapClass will meet the educational needs of both students and teachers learning computational thinking in their classrooms.

The Beauty and Joy of Computing (BJC) is an introductory computer science curriculum for AP CS Principles, a course developed by the College Board and the National Science Foundation. This course teaches students how to transform ideas into code using SNAP! (based on Scratch), one of the friendliest programming languages ever invented. But this course is far more than just learning to program. We focus on some of the "Big Ideas" of computing, such as abstraction, design, recursion, concurrency, simulations, and the limits of computation. We show some beautiful applications of computing that have changed the world, talk about the history of computing, and where it will go in the future.

Building on the foundations set by the AP Computer Science (CS) Principles course, this project seeks to dramatically expand access, especially for high school girls, to the most exciting and emerging frontiers of computing, such as distributed computation, the internet of things (IoT), cybersecurity, and machine learning, as well as other 21st century skills required to productively leverage computational methods and tools in virtually every profession. Creating pathways that stimulate high school learners' interest in advanced topics with the goal of building a diverse, gender-balanced, future-ready workforce is a crucial and impactful imperative addressed in this work.

Supported by DELTA

With DELTA at NCSU, we are experimenting with Moodle Analytics and intelligent dashboards to display important information for instructors in Moodle. We hope that these data visualizations and analyses will help instructors with student interventions, leading to better student course outcomes.

This project aims to serve the national interest in high-quality STEM education by studying an intervention designed to improve persistence of women in computer science. Engaging all people in learning about STEM, particularly computer science, is important for innovation, global competitiveness, and preparation for future careers that increasingly rely on computing. Women occupy just 28% of STEM jobs and account for only 17% of computer science majors and 21% of engineering majors. These differences in career choices arise partially from gender differences in self-assessment of STEM ability. This project will test whether educational institutions can use a simple intervention to increase the persistence of women in computer science. This intervention consists of email messages to students in introductory computer science courses. The messages contain contextual information about the student’s performance in class and encouragement regarding their effort and potential.

We are currently working on a few projects related to LLMs (particularly ChatGPT) in computing education spaces.

1. Teaching K-12 teachers how to use LLMs through an ongoing series of workshops.

This project consists of a series of workshops directed at K-12 teachers on how LLMs work, how they can be used in the classroom by students and teachers, and how to create an LLM-focused teaching lesson for students. These workshops have been run with RETs and Kenan Fellows.

2. Understanding the current state of literature for computing education.

This project is a systematic literature review of LLMs in computing education, which is an increasingly researched area. We are aiming to find identified strengths and weaknesses of these tools in the classroom, as well as advantages and disadvantages in their use.

3. Interviewing computing students, professors, and developers on how they use LLMs in their work.

This project entails interview studies to understand LLM usage among various computing professionals. At present, we have focused on interviewing professional, full-time software developers on how they use LLMs in their work, as well as analyzing the results. We intend to use these results to find best practices for students learning to program, as well as to understand how other groups (e.g., students) use these tools.

Diversity, Equity, and Inclusion

Our AI4SG team is interested in AI and machine learning research specifically with socially relevant applications. We have been exploring a wide variety of avenues by conducting a literature review, and currently are in the design process for our own ML/AI research study under the AI for Social Good topic. We are open to new students, and if you would like to learn more contact Lauren Alvarez and Zari Mcfadden.

Through the identification of appropriate tools and the co-creation of online and alternative "unplugged" remote learning projects that align with CS Principles learning objectives, our STARS CS for All Scholars program supports teachers in realizing their goals for providing more equitable access to CS learning experiences for high school students. University student and teacher teams will create activities and supports that allow for a range of CS activities, from totally unplugged for students with no computer and no internet, to computer-based for those with computers but no internet, to mobile-phone based, to online computer-based with collaboration support within the Snap programming environment. As the program evolves, STARS CS for All Scholars will conduct research on how to address challenges in fostering individualized and collaborative Computer Science Principles learning experiences.

The STARS (Students and Technology in Academia, Research, and Service) student organization at NC State University provides opportunities for undergraduate and graduate students to work together alongside NCSU faculty and staff on meaningful projects including research, community service, and outreach activities throughout the Triangle area. The STARS student organization is part of the STARS Computing Corps, a multi-institutional alliance with a mission to increase participation of students from underrepresented groups, and students with disabilities, in the discipline of computer science. We emphasize activities that both increase our university students' sense of identity within the computing discipline, and those that are likely to bring more students into the discipline in the future. Each month, members from the STARS organization host workshops for 2 concurrent FLAMES high school outreach programs (one within the MSEN program) and 2 concurrent SPARCS middle school outreach programs (one within the MSEN program). These workshops reach roughly 100 North Carolina students each semester. These outreach programs provide an opportunity for members to inspire young students to pursue computer science. In addition to these workshops, we partner with local schools and organizations (such as the Raleigh Police Youth and Family Services) for special community outreach events throughout the year including Hour of Code activities, lab tours, and panel discussions about participation in computer science.

The STARS Computing Corps is a national alliance with a mission to broaden participation of underrepresented groups in computing within institutions of higher education and consists of several initiatives. In particular, STARS aims to increase computing persistence and promote career advancement for undergraduates, graduate students, and faculty, with a focus on addressing systemic and social barriers faced by those from underrepresented groups in computing. The STARS Aligned research project seeks to investigate the impact of participation in STARS activities through analysis of alumni interviews, multi-year surveys, and other sources of data. The STARS Ignite program helps faculty prepare to support diverse cohorts of students to attend diversity-oriented conferences, providing scholarships and activities that build community, sense of belonging, and ignite efforts for diversity, equity, and inclusion within their own computing departments. Through workshops, webinars, and online resources, Ignite helps cohorts to follow up on what they've learned at the conference and through the Ignite program to design and implement projects that address their own local context and needs and to take action to broaden participation of underrepresented groups in computing in their home computing department.

Educational Games

"BOTs" is a block-based 3D puzzle game engineered as an educational platform to help newcomers acquire fundamental programming skills. In the game, players are given the mission of programming a robot to complete intricate puzzles that involve moving boxes onto designated switches. As players progress, the puzzles escalate in complexity while seamlessly integrating key programming concepts such as conditionals, loops, and functions. A distinguishing feature of "BOTs" is its focus on three-dimensional spatial navigation; the game allows players to manipulate the camera for a more nuanced understanding of each puzzle's architecture. Serving as more than just a game, "BOTs" functions as a compelling research tool, offering a rich environment for investigating spatial intelligence, problem-solving capabilities, and programming acumen.

Ongoing Projects

1. BOTs GUI Redesign (Development Project)

The primary objective of this project is to overhaul the existing BOTs' GUI, which currently relies on the ImGui toolkit-a platform with notable constraints in terms of efficiency and adaptability compared to contemporary GUI frameworks. To achieve this transformation, we intend to harness the robust capabilities and versatility of Unity's cutting-edge GUI system, along with state-of-the-art third-party assets, to entirely reimagine the BOTs' interface. The end result will offer players a modern, intuitive, and highly optimized user experience, elevating the quality of the gaming experience as a whole.

Skills needed: Unity Engine experience, C#, and Object-oriented programming.

Special Acknowledgement: Our redesign process will be greatly facilitated by the Blocks Engine 2 , a Unity asset developed by MeadowGames . We're grateful for MeadowGames' support in our endeavor.

2. Estimating Difficulty in 3D Programming Puzzle Games: The Role of Level Features and Spatial Skills (Research Project)

This study aims to understand how level features and spatial skills influence player performance and perceived difficulty in BOTs. We begin by examining the specific attributes of game levels that affect player performance and their sense of challenge. Next, we explore how spatial skills further modulate these effects. Our ultimate goal is to develop a predictive model for player performance that can inform personalized educational experiences in 3D programming puzzle games.

3. Co-op vs. Solo Play in Educational Programming Games: A Comparative Analysis (Research Project)

In this study, BOTS is used as a platform to investigate the relative effectiveness of cooperative multiplayer (co-op) and solo play modes in facilitating the learning of fundamental programming concepts. The goal is to determine which mode leads to better learning outcomes, providing insights for the design of future education tools for programming.

Have an Idea? Come talk to us. Venture IV, Suite 400.
You can see our legacy games on the archives page. Shelved Gaming Projects