Providing accessibility, AI-based teaching assistance, data-driven understanding of student progress!
For Students: learn and practice programming anywhere anytime
with only a Web browser
For Educators: teach effectively with in-depth visualization of student
progress driven by machine learning algorithms
For Universities: provide hands-on computer science education to
thousands of students
Challenges with providing CS education to a large number of students"
1.Hands-on programming exercises require special hardware and software configurations 2.Instructors have limited time to help students, but it is difficult to exactly pinpoint who needs help most and what kind of help 3. Students should learn from one another, but it is difficult to manage peer learning of programming outside of physical classrooms
Elices solutions to these challenges"
1. Provide a Web based platform that enables learning and teaching programming from anywhere anytime 2. Store, analyze, and visualize students real-time keystroke data to identify students in need of help and the specific points in the programs that they are stuck 3.Enable remote collaboration through screen sharing, synchronized peer editing and commenting (e.g., Google Doc), and 1-on-1 and group chatting
For Students: Accessibility"
A single online platform eliminates the need for
time-consuming installation and setup of a development environment!
Students can learn programming from !
anywhere at anytime using any device !
Students can easily access, review and study
past programming materials, and instructors monitor students codes and revisions; also provides online pair programming features!
For Instructors: Data-driven Management"
The online platform enables instructors to
efficiently teach and manage students at scale"
Instructors can analyze and visualize learning
data to determine overall progress (average scores, duration and number of submissions, revision history, etc.)!
Instructors can then supplement the course
curriculum based on students feedback, adjust the difficulty of assignments, identify and address common mistakes, and predict students progress !
For Students and Instructors:
AI Supported TA"
(Available Now) The platform gives faster and more
frequent feedback to individual students through features like automatic grading, code sharing, revision history review and plagiarism detection!
(On-going Work) Instructors can analyze and cluster
students data and identify similar groups, then determine which groups are doing well and which groups are in need of more assistance!
(Future Plan) Using AI and Machine Learning
techniques, the platform can help to identify frequently asked questions and provide automatic responses!
Leaders of Elice Project"
Alice Oh Faculty Advisor" alice.oh@kaist.edu"
Jae Won Kim"
Ph.D Candidate" jaewonk@kaist.ac.kr"
Research Interests: machine learning, text mining, computational social science"
Academic services include Program chair, (Senior) Program Chair for ACL, EMNLP, ACM KDD, ACM WWW, ACM WSDM, AAAI, AAAI ICWSM, IJCAI" 2008 Present, Assistant/Associate Professor, KAIST School of Computing" 2013 - 2014 Visiting Scholar, CRCS, Harvard University" 2008 Ph.D., Massachusetts Institute of Technology, USA" 2000 M.S., Carnegie Mellon University, USA"
Research Interests: applied machine learning, education technology"
2014- Present, Elice Project Lead" 2013 - 2014 Research Intern, CRCS, Harvard University" 2012 M.S. Dual Degree, KAIST, TU Berlin, Korea & Germany" 2009 B.S., University of Waterloo, Canada"