- Holder, D., Johnson, B, & Karol, P. (2002). A Consistent Set of Oxidation Number Rules for Intelligent Computer Tutoring.Journal of Chemical Education, 79, 465-467.
- Johnson, B., & Holder, D. (2002). A Cognitive Modeling Tutor Supporting Student Inquiry for Balancing Chemical Equations. The Chemical Educator, 7, 297-303.
- Walsh, M., Moss, C., Johnson, B., Holder D. & Madura, J. (2002). Quantitative Impact of a Cognitive Modeling Intelligent Tutoring System on Student Performance in Balancing Chemical Equations. The Chemical Educator, 7, 379-383.
- Johnson, B. & Holder, D. (2003). An Artificial Intelligence-Based Program for Automated Grading of Student Work in Balancing Chemical Equations. The Chemical Educator, 8, 271-279.
- Johnson, B. & Holder, D. (2010). A Model-Tracing Intelligent Tutoring System for Oxidation Number Assignment. The Chemical Educator, 15, 447-454.
- Kuhel, J., Wheeler, J., Miele, P., Holder, D., Johnson, B., Paterno Parsi, A., & Madura, J. (2010). Quantitative Impact of an Artificial Intelligence Tutoring System on Student Performance in Assigning Oxidation Numbers in Chemical Formulas. The Chemical Educator, 15, 455-460.
Developing and Validating Quantum Tutors: Supporting Anytime-Anywhere Learning
Murrysville, PA 15668
Quantum Simulations, IncMurrysville, PA15668
Projects Highlights - With support from the SBIR program at the Institute of Education Sciences (IES) in the U.S. Department of Education (ED) and other funding sources, Quantum developed a suite of web-based artificial intelligence (AI) learning tutors and assessment tools for chemistry and mathematics education. The tutors facilitate and scaffold learning through an interactive process that allows students to select their own problems, ask questions, and receive automated step-by-step tutoring feedback on their work. The web-based on-demand tutors are intended for K-16 students or adult or distance learners, and can be used in classrooms in direct support of a curriculum or as an independent tool for self-study, homework, or test preparation. Theuser interface is independent of the AI engine underneath, and can be modified to fit any age group.
To develop the product, Quantum used a tight iterative design-review-redesign process to ensure that the needs of users (e.g., students and teachers) and the education market were addressed. Design and development involved formulating procedural and conceptual approaches for each topic. For the tutoring component, a code library was created and a rule-based problem solver was implemented to enable the AI engine to accurately analyze student work and provide explanations to guide student work thereafter. Question-and-answer rules and content for student queries about problem solutions were developed and implemented. For the assessment component, detailed algorithms were created and implemented, and extensions to the AI framework were completed to accommodate assessment capabilities. An algorithm for automatically grading student work was also developed and implemented, along with a web-based interface to record that work. Finally, reporting capabilities were developed and implemented to create teacher reports on student process and concept mastery.
Field research conducted during these projects has demonstrated the feasibility of the tutors as they worked on all operating systems, and as teachers indicated the tutors can be integrated within their standard instructional practices. Research also demonstrated that students are engaged in learning when using the products. In one study of 304 high school students 92% stated they would recommend Quantum as a useful study tool. As well, results from a small pre/post pilot study of 51 high school students showed an improvement of more than a full letter grade for students in the treatment group (N=31) compared to a business as usual control group (N=20). Additional research on the impact of the Quantum chemistry tutors to improve student learning is currently underway through an efficacy trial funded by the Institute’s Math and Science Research Grants Program. The study is being conducted in California by WestEd, an independent evaluator. The study is a randomized controlled trial in which forty-five high school teachers in schools assigned to the treatment group will teach their usual curriculum with Quantum’s 12 chemistry tutors and 45 teachers in schools selected for the control group will use only their usual curricula. Results will likely be available in 2012.
Commercialization Path - Quantum has commercialized its tutors through direct sales, as well as distribution partnerships with established organizations such as textbook publishers Holt, Rinehart and Winston (part of Houghton Mifflin) and McGraw-Hill. Partners are able to incorporate Quantum in various and distinctive ways such as integrated with homework systems and online learning materials. As of 2011, Quantum has received $1.5 million in integration/licensing revenue, has delivered more than half a million online tutoring sessions, and presently has more than 10,000 users in high schools and colleges nationwide.
Peer reviewed publications resulting from this R&D
Intellectual Property - Intelligent Tutoring Methodology Using Consistency Rules to Improve Meaningful Response, B. G. Johnson and D. A. Holder; Patent Number US 6,540,520 B2, issued April 1, 2003.
About Quantum Simulations - Dedicated to its mission to improve education, Quantum Simulations, Inc. is a leader in intelligent technologies and is on the cutting edge of innovation. Quantum develops unique artificial intelligence (AI) tutoring, assessment and professional development software for accounting, science and mathematics that empowers instructors and inspires students in K-16, adult learning and distance education programs to achieve more.
Company website: http://www.quantumsimulations.com/index.php
Product Demo: http://www.QuantumTutors.com/demo.html
IES SBIR Program website:http://ies.ed.gov/ncer/sbir/