Research Intern
at IEC
The Innovative Educational Computing Lab is hiring a few graduate students (MS and PhD) as research and software engineering interns on the following projects. The appointment will be either as GRA (at a contracted monthly stipend with tuition benefit), RA (at a per-hour rate payment without a tuition benefit), or volunter (no pay). Interested candidates should email Dr. Matsuda (Noboru.Matsuda@ncsu.edu) a current CV and a statement of research indicating the project of interest.
Project 1: Interactive Map for History Education Project [Full-stack programmer]
The goal of the project is to develop an interactive map comparison tool that allows students to compare modern and ancient maps. We hypothesize that the interactive map comparison tool will help students understand how individual countries developed their perception of other countries, which in turn facilitate student's learning to understand how the concept of nation and modern international relationships have evolved.
An ideal candidate would have at least one year of industry experience as a full-stack programmer with the following skill set: React, Express, Node.js, MongoDB or some other Database (can be SQL). Experience working with Geographic Information Systems would be preferred.
Project 2: Adaptive Online Courseware Project [Full-stack programmer]
The goal of the project is to develop a web-based platform to create and host adaptive online courseware. The adaptive online courseware has been deployed on existing open-source system, Open edX. We have extended Open edX to a great extent by integrating several AI techniques (e.g., integration of intelligent tutoring systems, analysis of courseware contexts to optimize assessment sequencing, etc).
An ideal candidate should have a solid record in full-stack Web developer with expertise in Python, Java, Django, REST, JavaScript, JSP, HTML, and CSS. Knowledge and experience on the Linux OS, Docker, and Tomcat technologies would be ideal. Spell our your experience as a full-stack Web developer when you apply.
Project 3: Online Platform for Learning by Teaching [Full-stack programmer]
The goal of the project is to develop a web-based app for learning by teaching (www.SimStudent.org). The primary task is to convert a Java applet version of our research intervention, called APLUS, into a web app. APLUS allows students to interactively teach a synthetic peer, called SimStudent, how to solve linear equations, but the theory conjectures that it is actually the student who learns by teaching. In other words, using the APLUS app, we investigate how and when students learn by teaching.
An ideal candidate for the full-stack programmer should have a solid record in full-stack Web developer with expertise in Java, REST, Spring Boot, Angular, SQL, JavaScript, JSP, HTML, and CSS. Knowledge and experience on the Linux OS and Docker technologies would be ideal. Spell our your experience as a full-stack Web developer when you apply.
Project 4: Sequence patten mining for cancer detection and treatment [Machine Learning]
The goal of this project to develop a machine learning model to analyze DNA sequences to understand how particular type of breast cancer develops to further develop a reliable system for early detection and treatment. This is a joint collaboration with a professor in medicine at the University of North Dakota.
Ideal candidates should have in-depth experience and theoretical understanding in deep learning and sequence mining beyond course projects.
Project 5: Video analysis for anomaly behavior detection [Deep Learning on Video Analysis]
There is growing demand for automated surveillance in broad domains. One of the critical tasks is to detect anomalous behavior captured in video recordings. We apply deep neural network technology to build a highly reliable model for anomaly detection. Although, the technology is (or should be) task independent, we are currently interested in animal behavior, in particular bred animals in a farm (e.g., detecting sick animal).
Ideal candidates should have in-depth experience and theoretical understanding in deep learning beyond course projects.
Project 6: NLP analysis for scalable online course engineering [Deep Learning on NLP]
We are looking for MS students for various tasks on the PASTEL project (www.ieclab.org), including NLP applications for question generation and course contents analysis, reinforcement learning applications for optimal sequencing of instructions and quality assurance of the courseware contents. The actual project task will be identified based on the skills and interests of the candidates.
Ideal candidates should have in-depth experience and theoretical understanding in artificial intelligence and machine learning (any particular techniques) beyond course projects.
Project 7: Automatic generation of pedagogically profitable questions for adaptive MOOC [Deep Learning on NLP]
The goal of this project is to develop an NLP text-generation system that automatically generate questions with high pedagogical utility from instructional text (e.g., online textbook). The generated questions will be then integrated into the adaptive online courseware, called Cyberbook, a kind of massive-open online course (MOOC). We currently hypothesize that pedagogically profitable questions can be generated by analyzing relationship between learning objectives and instructional texts. As a proof of concept, a regression-based model that learns a latent association between given learning objectives and instructional texts has been developed. Once the latent association is learned, a representative sentence(s) in the instructional texts will be converted into a question. This is an NSF funded project called PASTEL. Visit our web for details: www.ieclab.org
An ideal candidate should have a decent experience on a research project with strong theoretical background in deep learning, machine learning, and natural language processing.
Project 8: Optimal sequencing for adaptive online courseware to facilitate robust learning of interconnected knowledge [Reinforcement Learning]
The goal of this project is to develop a computational model for optimal sequencing of instructions implemented on adaptive online course (aka adaptive MOOC). One of the challenges when learning a complicated subject is a lack of in-depth understanding of the complex web of knowledge where many different kinds of knowledge are involved. We hypothesize that one way to facilitate learning such complex web of knowledge is to provide students with rich multimodal instruction that consists of mixed representations and adaptive scaffolding. How to best sequence such a rich set of instructions becomes a problem. We propose to apply reinforcement learning to dynamically compute optimal sequence of instruction for individual students. This is an NSF funded project called PASTEL. Visit our web for details: www.ieclab.org
An ideal candidate should have a decent experience on a research project with strong theoretical background in reinforcement learning, machine learning, and natural language processing.
Project 9: Studying the effect of learning by teaching a synthetic peer [Machine Learning for Science of Learning]
The Innovative Educational Computing Lab (www.ieclab.org) is looking for a PhD student to conduct a research on an externally funded project, called SimStudent, where we study how students learn by teaching (www.SimStudent.org). We apply an interactive machine-learning technology to deploy a pedagogical agent that serves as a synthetic peer for a student to teach. Potential thesis projects include (but not limited to) developing a dialogue model to facilitate the effect of learning by teaching, extending the built-in machine learning model of the synthetic peer to deal with multimodal instructions (e.g., written texts and diagrams), conducting school study to advance a theory of learning by teaching. This is an NSF/IES funded project called SimStudent. Visit our web for details: www.SimStudent.org
An ideal candidate should have a decent experience on a research project with strong theoretical background in artificial intelligence, machine learning, and learning science.