We have two positions available for undergraduate interns for the summer of 2010 supported by National Science Foundation, Research Experience for Undergraduate.
Title: Learning by Teaching a Synthetic Student -- A study of the effect of tutor learning using SimStudent as a teachable agent
Project Description: The primary purpose of the current SimStudent project (www.SimStudent.org) is to study cognitive and social theories on the effect of tutor learning -- a well-known effect that students learn when they teach their peers. To study how and when the effect of tutor learning happens, we are building an on-line game-like learning environment where students learn how to solve algebra equations by teaching a live computer agent, i.e., SimStudent.
The SimStudent architecture and the proposed learning environment provide a rich study environment where we can conduct a various controlled studies to explore the cognitive and social factors on the effect of tutor learning. There would also be opportunities to explore issues in machine learning such that improving SimStudent’s learning mechanism, or issues in human-computer interaction such like interactions between student’s and SimStudent. For potential summer intern projects, we will provide each individual intern an independent research project based on his/her interest and experience. There are a number of potential projects that would be suitable for a summer intern projects as listed below (not exclusive).
(1) Educational Data-mining Study – How do students learn by teaching? We plan to have an evaluation study in algebra I classes in two local high schools with the total of about 100 students involved in Spring of 2010. By the time an intern starts the project, a large amount of students’ learning activity log data will be available at so-called DataShop, an open research data repository maintained by Pittsburgh Science of Learning Center. Thus, a primary goal of a potential research project would be to analyze these rich dataset to explore cognitive and social factors that mediate the tutor learning.
(2) Usability Study – What makes the Learning-by-Teaching environment user-friendlier and hence facilitate the tutor learning more? From the human-computer interaction point of view, the improvement of the systems’ usability is an essential key for the success in accomplishing our research agenda. In this potential research project, the intern student will apply various HCI methods to evaluate the system’s usability and explore the key HCI factors to maximize the students’ learning.
(3) Prior Knowledge Study – How does the “individual” differences of SimStudent (i.e., the tutee) affect the student’s (i.e., the tutor’s) learning? By manipulating the background knowledge of SimStudent, we can control the speed and accuracy of SimStudent’s learning. For example, SimStudent may start with a certain amount of knowledge for equation solving, or SimStudent might have immature or even irrelevant background knowledge that slows down learning rate and causes more errors. The goal of the intern project would then to study how such differences affect the tutor learning.
(4) Tutoring Dialogue Study – How could we improve interaction between SimStudent and the students? So far, SimStudent only learns from the steps demonstrated. It would be more natural and (perhaps) pedagogically more appropriate if the student (the tutor) could give a hint with his/her own everyday language (e.g., “you can subtract the same number from both sides”). Such a natural language input could be used as a heuristic to navigate the search for induction. The goal of the intern project here is to study a rich tutoring dialogue by implementing and testing an augmented dialogue facility in the learning-by-teaching environment.
(5) Machine-Learning Study – Can we improve the SimStudent’s learning algorithm? So far, we used inductive logic programming that is basically implemented as a brute forth search. We also use FOIL (Quinlan, 1990), which learns Horn Clauses from relations provided in examples. There are pros and cons for the current implementation (mostly the issues for domain generality and knowledge representation). The intern on this project would study alternative technologies to enhance the generality and efficiency of SimStudent’s learning.
(6) Cross-domain Generalization Study – Can we apply to another domain? So far, we have implemented the Learning-by-Teaching mostly on the math domain (equation solving and quadratic functions). A further investigation on the generality of the proposed architecture is needed. The goal of this project is to study a generality of the SimStudent technology to build a Learning-by-Teaching environment for different subject domains.
Duration: 10 weeks in any time from June 1 to August 31
Eligibility: The applicants must be undergraduate students and U.S. citizens (or U.S. permanent resident).
Contact: Noboru Matsuda <Noboru.Matsuda@cs.cmu.edu>
Human Computer Interaction Institute
Carnegie Mellon University
5000 Forbes Ave. Pittsburgh, PA 15213
Voice: 412-268-2357 Fax: 412-268-9433