Research Projects

Active Projects

Camp DIALOGS: Middle School AI Summer Camps

In the Camp DIALOGS project (project website), we provide two-week AI summer camp opportunities to middle school students living in Gainesville, Florida. We support students’ AI learning through a variety of lessons and activities, including collaboratively building conversational agents (like Siri or Alexa). To better support students’ creation of conversation AI, we design and build an interface, AMBY (AI Made By You). To understand the design needs, features and interface usability, we conduct participatory design for AMBY with the middle school students throughout the year and iteratively refine the interface. Using AMBY, students can add intents, training data and responses to build their own conversational AI. AMBY will be deployed beginning the summer of 2022.

PRIME: Engaging STEM Undergraduate Students in Computer Science with Intelligent Tutoring Systems

This is a collaborative research project at UF and NC State University (project full description here). PRIME is an adaptive learning environment designed to support STEM undergraduates in learning computer science concepts through block-based programming. PRIME utilized Google’s Blockly framework to support block-based programs. We created adaptive, multilevel hints and feedback to help students on their computational thinking and problem-solving. In addition, we built student models about their affective and cognitive states, which we hope to inform when and how the system should offer intervention to students while completing the activities.

Linguistic Alignment in Collborative Coding Tasks

Collaborators: Amanda Griffith, Zane Price
University of Florida, Computational Linguistics course project

Linguistic alignment arises when interlocutors begin to share common linguistic features during conversation. This phenomenon has shown to be predictive of task success, and to vary by age groups. Pair programming is a collaborative task that has various benefits in computer science education. Investigating linguistic alignment in pair programming tasks can help understand the key aspects of coordination and inform design of adaptive learning environments. In this study we investigate the low-level linguistic alignment (syntactic and lexical) in two pair programming corpora from 36 pairs of middle school and 15 pairs of college students. The results show that middle school students has higher syntactic alignment and lower lexical alignment than college students. More detail can be found in our [poster].

Past Projects

StudyBuddy: A Chatbot for Effective Study Habit Behavioral Change

Collaborators: Ishrat Ahmed, Arun Balajiee, Zak Risha, Jacob Biehl
University of Pittsburgh, School of Computing and Information

Check out our 3-minute prototype demo!

This is a course project for CS 3570 Advanced User Interface Seminar. In the transition to a new stage of learning, college freshmen are in specific need of developing their study habits and skills to achieve independent work in higher education. We designed a chatbot StudyBuddy to support freshman’s behavioral change. We administered interviews with peer tutors and surveys freshmen at Pitt’s CS department, which we found students have difficulties managing their project and time. we deployed StudyBuddy in Slack, that periodically sends tips, provides assessment of students’ study habits via surveys, helps the students break down assignments, and sends reminders. We finally offered design guidelines of the chatbot supporting learning behavioral change for college Computer Science students.

Rapport Management in Multi-session Interactions with a Social, Teachable Robot

Collaborators: Nichola Lubold, Leah Friedman, Erin Walker
University of Pittsburgh, Facet lab

This is an independent research project. I studied middle school students’ rapport (interpersonal closeness) with a robot who speaks socially called Emma for multiple sessions. Prior research has investigated the effects of social robots on student rapport and learning in a single session, but little is known about how individuals build rapport with a robot over multiple sessions. We modeled learners’ rapport-building linguistic strategies to understand whether the ways middle school students build rapport with the robot over time follow the same trends as human conversation, and how individual differences might mediate the rapport between human and robot.

Allo Alphabet: Mobile Literacy System Improving Children’s French Literacy in West Africa

Collaborators: Michael Madaio, Amy Ogan
Carnegie Mellon University, Human Computer Interaction Institute

This is my research intern work at Carnegie Mellon. Allo Alphabet is an IVR and SMS based literacy system that is currently deployed in rural Côte d’Ivoire. Literacy has been linked to pervasive poverty, unemployment and illness, educational technologies can help mitigate low levels of childhood literacy. But some children may get developmental delays in pre-literacy skills due to their family’s literacy level, bilingual environment, etc. Given that computer-mediated literacy learning systems often have a fixed progression through the curriculum, it might not effectively support different learners. My research involves modeling students’ phonological awareness skills using Bayesian Knowledge Tracing and investegating factors that associate with differing the learnabilities of children’s pre-literacy skill. With our modeling, we can provide more adaptive support for different individuals in the literacy learning systems.

Social Media Attention Effectiveness for Fundraising among Nonprofit Organizations

Collaborators: Rosta Farzan
University of Pittsburgh, Sustainable Social Computing lab

This is an independent study supervised by Dr. Rosta Farzan during Spring 2019. I explored how nonprofit organizations’ use of social media for publicity and for gaining donations. I analyzed 414,312 Twitter posts of local non-profit organizations in Chile and their donations records. I found that a nonprofit being “heard” does not solely depend on how “loud” it speaks (i.e. the number of tweets), instead, these accounts also rely on “whom” they speak to and “where” they speak (i.e. Mentions “@” and Hashtags “#”). Using the findings from this research, nonprofit organizations can allocate their restricted capacity more effectively.

Online Educational Information Quality Modeling and Perceived Difference Comparison

Collaborators: Jing Li
Anhui University, Department of Management Science

This is my undergraduate research training project advised by Dr. Jing Li at AHU. We investigated the relationship between online educational information resource quality and learning performance. We conducted 233 questionnaires, utilized structural equation model (SEM) to verify our proposed theoretical model. We found that the content quality, form quality and utility quality of the online educational information resources had positive impact on information usage, and then positively influenced the user’s learning performance. We conducted qualitative analysis on differenct information seeking behaviors among online courses (such as MOOC), search engine and education Q&A platforms (like Zhihu), which offered design implications for improving learning experience of people using online educational resources.