Zhongxiu Aurora Peddycord-Liu

Publications Work&Projects Hobbies CV

MIND Research Institute: ST-Math 2016-present

ST Math, developed by MIND Research Institute, is an educational game for K-12 and is designed to boost math comprehension and proficiency through visual learning. ST-Math is a curriculum-integrated program currently used by over a million students. ST Math is identified by the CEO-led Change the Equation as one of just four STEM programs ready to scale nationally.

I am the lead student on NSF-funded project (Evaluation for Actionable Change: A Data-Driven Approach), in collaboration with Dr. Rutherford from NC State College of Education, Dr. Lynch, and my advisor Dr. Barnes. This is NC State's first project on ST Math. I initiate, design, and execute research projects; I collaborate across disciplines and mentor other PhD students.

  • I apply and create data-mining and analytical methods to understand gameplay and learning, to evaluate games against their educational goals, and to derive actionable game-design feedback.
  • I employ qualitative approach to understand the use of ST Math in classroom context, and to inform game design that benefit practioners.

Game2Learn Lab: BOTS 2015-present

BOTS is a programming puzzle game designed to teach fundamental concepts of coding to novice programmers. In the game, players write code in a drag-and-drop programming language. The object of the game is to move a robot character around mazes, solving puzzles and clearing obstacles using programming concepts such as control flow and function. Aurora's research has been on:

  • Debugging in BOTS. How do students' programming experience and problem solving behaviors related to debugging performance?
  • How level-creation tool designs affect the quality of player-generated content (in this case, the maze puzzle).

Game2Learn Lab: Deep Thought 2014-2015

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. Aurora's research has been on:

  • When and how to combine worked example with problem solving?
  • Data-driven method to send students to easier/harder problem paths based on their proficiency profile.

WPI AI Lab: ASSISTments 2012-2014

ASSISTments is an intelligent tutoring system developed at Worcester Polytechnic Institute and used by tens of thousands of users worldwide. Aurora's work has been on:

  • Implemented a features that enable teachers to design the content (message, media) and condition (affective state detected by sensor-free detectors, students correctness in a row) that will trigger intervention messages.
  • General programming, trouble-shooting, and assisting in classroom experiments.

Hungarian Academy of Sciences: Automated Chinese-Hungarian Dictionary Spring 2013

MTA-SZTAKI is a research institute, well-reputed for automatically creating large-scale online dictionary service. Aurora did her undergraduate major qualification project by creating a pipeline of building a sentence-level parallel corpus and an automatic bilingual dictionary between Chinese, a high-density character-based language, and Hungarian, a medium-density word-based language.

SAS Education Analytics Summer 2017-present

I am a year-long intern working in SAS Advanced Analytics Education team.

  • I analyze SAS e-learning data to derived actionable feedback for course design. I presented my work at the education department's mid year summary, at executive meetings, and was interviewed here.
  • I participated in technical review and test teach of courses that use SAS tool for data analytics and predictive modeling.
  • I co-developed SAS Advanced Analytics Education's "Neural Work: Essential" course

IBM Waston for Oncology Summer 2015

IBM Watson is the Jeopardy winning cognitive system that uses natural language processing and machine learning to reveal insights from large amounts of unstructured data. Watson for Oncology is Watson's application on healthcare, which provides physicians with evidence-based treatment options from large volume of research, medical records, and clinical trials. Aurora's internship contributed:

  • Collaborated in a team of 5, designed and implemented visualization to compare treatments' side-effects while adaptive to patients' personal preferences.
  • The design process actively involved customers and designers. The product follows principles of cognitive computing, intuitive, pluggable, and friendly to visually impaired users.
  • Applied Agile development and Angular JS.
  • Resulted in a patent in submission.

Dell Summer 2013

I worked with Dell's Equallogic Storage Solution team. I automated the procedure to evaluate disk drive performance. The procedure evaluates various disk controllers and models, and generates statistical report to help the team make informed decision on choosing storage disk.

SPARCS Middle School Outreach Fall 2014-Spring 2017

I lead STARS Computing Corps' SPARCS MSEN program - the NC Math and Science Education Network where SPARCS is offered as an elective course. I organize and design courses to teach 6-8 graders computer science concepts and programming, using interative tools such as snap, gamemaker and BOTS




Flag Counter

Last update Aug 2018 by Zhongxiu Liu