IMPORTANT UPDATE ON COVID-19
August 4, 2020: In light of developments surrounding COVID-19 pandemic, and for the health and safety of our patrons and staff, we are abiding by the CDC's recommendations of canceling gatherings for the next several weeks. See UMass Amherst's Fall 2020 reopening plan here.
The Center for Knowledge Communication urges everybody to practice social distancing measures.
We are recruiting for the WestEd PALS study. Read our latest announcement here. If you have any questions regarding the study, please contact us.
The Center for Knowledge Communication
Investigating knowledge-based educational systems, integrating theoretical principles into research systems for empirical evaluation and theoretical analysis.
Education Is Not a "One Size Fits All" Proposition
The Center For Knowledge Communication Addresses A Growing Need In Education
Our focus is on tailoring activities to meet the individual needs of a diverse student population. The challenge is to meet these needs in a traditional classroom.
Though teachers are responsible for more students than before in the classroom and online, it is possible to produce a private computer tutor for each student and increase the teacher's ability to respond to each individual.
Research Professor Beverly Park Woolf is building Web-based intelligent tutors that understand a student's learning needs, optimize teaching materials, and use effective tutoring strategies. "These tutors, when used in traditional classrooms with caring teachers, provide the advantages of individualized instruction at an affordable cost," says Woolf.
Machine tutors operate like a trusted mentor, speeding through topics that the student grasps easily, concentrating on topics that cause trouble, and never losing patience.
Some research challenges include:
How does a computer tutor learn to adapt to each student's learning style and cognitive skills?
How does a computer track student collaboration and identify each student's contribution?
How does a tutor help unmotivated students to focus on learning?
Behavioral Studies and Student Modeling
Intelligent tutors are built on models of knowledge that represent the key ideas to be learned, common misconceptions, and how a student's knowledge changes over time. Modeling an individual's knowledge is challenging since students' knowledge is often confused or incomplete.
Dr. Woolf has built many intelligent tutors in collaboration with colleagues in psychology, chemistry, engineering, biology, ecology, geology, education, and medicine. Some of the most promising tutors were built with Research Scientist Ivon Arroyo and included tutors for elementary and high school mathematics.
The effects on individual learning differences (e.g., mathematical ability) and group characteristics (e.g., gender) were documented by deploying two tutors - AnimalWatch (arithmetic) and Wayang Outpost (geometry) - among nearly a thousand elementary and high school students.
Large-scale experiments determined the practical significance of each tutor and showed that they have a measurable impact: students who used these tutors for two or three hours improved their results on Massachusetts-required standardized exams by 10-12%.
These multimedia tutors contain nearly 1000 problems and are supplemented with data about cognitive features of each student, including variables for individual differences. For example, machine learning was used to modify tutor behavior according to each student's Piagetian developmental stage, spatial ability, or math-fact-retrieval skills.
Evaluation results indicate that students with low cognitive skills learn best with concrete representations and manipulatives, and those with higher cognitive skills learn best with abstract or symbolic representations. The effects of gender characteristics in learning have also been measured. Female students spend about 25% more time on hints than male students, perceive a tutor more positively than male students, and are more willing to use the tutor again. Boys with low cognitive development perform worse when they receive abstract or symbolic help while boys with advanced cognitive skills seem to learn better with abstract help than with hints.
While using the computer tutors, students with weak skills benefit the most, seem comfortable requesting hints, make use of help and instruction, and demonstrate improved performance. This is the reverse of the usual findings in the classroom, where higher achieving students are most likely to request help.
Learning to Teach
Machine learning (ML) techniques are used to model each student's skills and to optimize the selection of problems and hints. During tutoring sessions, the tutoring program can assess a student's skills by considering variables such as prior knowledge and the level of a student's engagement in the tutoring process. Bayesian and data mining techniques help identify a student's skills and predict student reactions to a variety of teaching styles (e.g., present a hint or an example) and to understand how each student learns.
Bayesian nets are used to reason about a student's affective state (motivation, engagement, interest, and learning) and to discover links between observable behavior (time spent on hints, number of hints selected) and hidden variables (attitudes and goals). Correlation between observable student activity and survey responses are converted into a network that tests the predictions on the data log of new students.
"Using ML techniques, we can predict a student's level of engagement with 80-90% accuracy, and how a student will perform on each problem with 70% accuracy," says Woolf.
Web-enabled intelligent instruction
Woolf is developing several tutors to personalize Web instruction based on presumed student knowledge, cognitive skills, and learning needs. One intelligent tutor can learn in just a few Web pages how to classify a student's learning needs and which teaching approach to use, on a per-student basis. This tutor adapted each slide of a Web-enabled lecture course based on its prediction about which features a student would most like to see (e.g., definition, explanation, example). The tutor examined each topic at several difficulty levels and used a Naive Bayes Classifier algorithm to determine whether or not the topic should be taught and to manage the layout of each page.
Beyond traditional classrooms
Students are often passive in classrooms; they are not regularly involved in thinking, active learning, problem-solving, or argumentation. In the traditional classroom, teachers ask 95% of the questions, mostly requiring short answers. Traditional classroom methods - lectures, books, multiple-choice exams - lead to passive students and are successful only with the top 25% of students. Liberal use of interactive graphics (3D modeling and interactive character animation) and sound within intelligent tutors help teachers connect with all students.
Woolf says "We do not intend for this technology to be used to imitate conventional classroom approaches; rather we focus on challenging traditional teaching and supporting new teaching methods." Intelligent tutors play an essential role in moving education towards more student-centered methods, e.g., team collaboration, case-based inquiry, and apprenticeship; techniques that are nearly impossible to implement without technology.
"As cognitive science and psychology continue to broaden our understanding of how people learn, a real possibility exists to produce a teacher for every student," says Woolf. Thus, content, teaching, assessment, student-teacher relationships, and even the concept of educational institutions may all need to be rethought.
Made with ❤️in University of Massachusetts Amherst College of Education
Woolf is developing several tutors to personalize Web instruction based on presumed student knowledge, cognitive skills, and learning needs.
One intelligent tutor can learn in just a few Web pages how to classify a student's learning needs and which teaching approach to use, on a per-student basis. This tutor adapted each slide of a Web-enabled lecture course based on its prediction about which features a student would most like to see (e.g., definition, explanation, example).
The tutor examined each topic at several difficulty levels and used a Naive Bayes Classifier algorithm to determine whether or not the topic should be taught and to manage the layout of each page.
White House Names UMass Computer Scientist Beverly Woolf a Presidential Innovation Fellow
Effort-Based Tutoring: An Empirical Approach to Intelligent Tutoring. Ivon Arroyo, Hasmik Mehranian, Beverly Park Woolf.
The research team of the Center for Knowldge Communication is proud of the fine work done by Ivon Arroyo, David G. Cooper, Winslow Burleson, Beverly Park Woolf, Kasia Muldner and Robert Christopherson whose research paper and related presentation won top awards at the 14th International Conference on Artificial Intelligence in Education (AIED 2009)
Woolf, B., Ghosh, A., Lan, A., Zilberstein, S., Juravich, T., Cohen, A., Geho O., (2020).
In AAAI Spring Symposium 2020: AI in Manufacturing, Stanford University.
Joshi, A., Allessio, D., Magee, J., Whitehill, J., Arroyo, I., Woolf, B., Sclaroff, S., Betke, M. (2019).
In: 13th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition, FG 2019.
Allessio, D., Woolf, B., Wixon, N., Sullivan, F., Tai, M., Arroyo, I (2018).
Artificial Intelligence in Education. Springer International Publishing. London, UK.
Wixon, N., Woolf, B.P., Schultz, S., Allessio, D., & Arroyo, I. (2018)
Springer International Publishing. London, UK.
Arroyo, I., Wixon, N., Allessio, D., Woolf, B. P., Muldner, K., & Burleson, W. (2017).
Springer International Publishing. Wuhan, China.
Karumbaiah, S., Lizarralde, R., Allessio, D.,Woolf, B., Arroyo, I., (2017)
Arnon Hershkovitz, Luc Paquette, (Eds) 10th International Conference on Educational Data Mining. Wuhan, China, Springer International Publishing.
Wixon, N., Schultz, S., Muldner, K., Allessio, D., Burleson, W., Woolf, B., & Arroyo, I. (2016, July)
In Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization (pp. 311-312). ACM.
Wixon, N.; Schultz, S., Muldner, K., Allessio, D., Woolf, B.P., Burleson, W., Arroyo, I. (2016)
The 13th International Conference on Intelligent Tutoring Systems. Zagreb, Croatia, 2016.
Schultz, S. E., Wixon, N., Allessio, D., Muldner, K., Burleson, W., Woolf, B., & Arroyo, I. (2016)
International Conference on Intelligent Tutoring Systems, 314-319. Springer.
Wixon, Allessio, D., Ocumpaugh, J., Woolf, B., Burleson, W., Arroyo, I. (2015)
International Workshop on Affect, Meta-Affect, Data and Learning (AMADL 2015), 34-44.
Why do you feel that way? Exploring Students’ Attributions for Emotions in MathSpring ITS
Wixon, N., Allessio, D., Muldner, K., Schultz, S., Burleson, W., Woolf, B., Tapia, S., Arroyo, I. (EDM, 2015).
Paper presented at EDM 2015 conference.
Woolf, B. P., & Arroyo, I. (2015).
IBM Journal of Research and Development, 59(6), 9-1.
Woolf, B.P., Lane, H.C., Chaudhri, V.K., Kolodner , J.L., (2014).
AI Magazine. Special issue on Intelligent Learning Technologies: Applications of Artificial Intelligence to Contemporary and Emerging Educational Challenges (V. Chaudhri, D. Gunning, H.C. Lane & J. Roschelle, Eds). pp 68 - 84.
Arroyo, I., Woolf, B.P., Burleson, W., Muldner, K., Rai, D., Tai, M. (2013).
International Journal on Artificial Intelligence and Education, Special Issue on Landmark Learning Systems (B. McLaren, S. Sosnovsky, Eds.).
Ivon Arroyo, Beverly P. Woolf, David Cooper, Winslow Burleson, Kasia Muldner International Conference of Advanced Learning Technologies (ICALT 2011) Athens, Georgia, July 2011.
Woolf, B.P., (2009).
Elsevier Publishing Morgan Kauffman, San Francisco, CA.
Minghui Tai, Beverly P. Woolf, Ivon Arroyo International Conference of Advanced Learning Technologies (ICALT 2011) Athens, Georgia, July 2011.
Imran A. Zualkernan, Ivon Arroyo, Beverly P. Woolf International Conference of Advanced Learning Technologies (ICALT 2011) Athens, Georgia, July 2011.
Ivon Arroyo, James M. Royer, Beverly Park Woolf International Journal of Artificial Intelligence in Education. Selected Best Papers from ITS 2010 conference.
Shanabrook, D., Cooper, D., Woolf, B., Arroyo, I. - Proceedings of the Third International Conference on Educational Data Mining, June 11 - 13, 2010 - Pittsburg, PA
Arroyo, I., Mehranian, H., Woolf, B. 2010 Best Paper Award for 3rd International Conference on Educational Data Mining - June 11 - 13, 2010 - Pittsburg, PA. Presentation.
David G. Cooper, Kasia Muldner, Ivon Arroyo, Beverly Park Woolf, Winslow Burleson - Submitted to the International Conference on User Modeling and Adaptive Presentation, 2010, Hawaii.
Arroyo, I., Cooper, D.G., Burleson, W., Woolf, B.P., Muldner, K., Christopherson, R. (2009).
International Conference on Artificial Intelligence and Education, IOS Press.
A Data Mining Approach to Intelligent Tutoring
Ivon Arroyo, Hasmik Mehranian, Beverly P. Woolf. Journal of Educational Data Mining (JEDM).
Beal, C. R., Arroyo, I., Cohen, P. R., & Woolf, B. P. (2010) Journal of Interactive Online Learning.
Woolf, B.P., Arroyo, I., Muldner, K., Burleson, W., Cooper, D., Dolan, R., Christopherson, R.M.
Ivon Arroyo, Beverly P. Woolf, James Michael Royer, Minghui Tai, Kasia Muldner, Winslow Burleson, David Cooper - Technical report UM-CS-2010-020. Computer Science Department, UMASS Amherst
IN THE NEWS
Beverly Park Woolf PhD is a Research Professor in the Computer Science Department of the University of Massachusetts Amherst. She is the Director of the Center for Knowledge Communication. Many of the three-dimensional graphics and multimedia classes at the University of Massachusetts owe their beginning to Dr. Woolf's efforts to offer students the opportunity to expand both intellectual and practical skills.
Dr. Woolf's research focuses on building systems to train, explain, and advise users effectively. Extended multimedia capabilities are integrated with knowledge about the user, domain, and dialogue to produce real-time performance support and on-demand advisory and tutoring systems. The tutoring systems use intelligent interfaces, inferencing mechanisms, cognitive models, and modifiable software to improve a computer's communicative abilities. These systems have been tested with learners, trainers, and other client bases, deployed in education and industry and demonstrated in more than 50 American industrial, military and academic sites and 15 foreign countries.
Her most recent book is Building Intelligent Interactive Tutors, Student-Centered Strategies for Revolutionizing E-Learning, Published by Elsevier & Morgan Kaufmann, 2008.
For more information about past and present multimedia classes offered by the Center for Knowledge Communication, please use our contact form.
Dr. Arroyo holds a Licenciatura en Informatica Universidad Blas Pascal - Computer Science (1995) - (equivalent to Bachelors of Science) Cordoba, Argentina - an M.S. from the University of Massachusetts Amherst in Computer Science (2000) and an Ed. D. from the University of Massachusetts Amherst (2003).
She is a Fulbright Fellow and an elected member of the executive committee of the International Society of Artificial Intelligence in Education. Her current work focuses primarily on Wayang Outpost a geometry tutor for middle and high school students.
In her interdisciplinary role, she has carried out top research at the forefront of education, computer science, and psychology, co-authoring research articles at the forefront of the three disciplines.
From the education perspective, she has researched and created learning software for mathematics with multimedia, and worked closely with thousands of K-12 students and teachers, deployed software in public schools, while trying understand how students best learn and perceive mathematics with interactive math tutoring software, and how to support teachers in their teaching process via digital assessment tools.
From the computer science perspective, she has created artificially intelligent tutoring software that models students' knowledge and affective states, infers them from student behaviors and physiological sensors, and which acts upon those students states, as students use the software. She has used data mining methods to learn a variety of student states from past student data logs.
From the psychology perspective, she has analyzed developmental gender differences in the use and benefit of math tutoring software, cognitive development issues in relation to the best representations to use while teaching mathematics, and memory retrieval studies where the training of speed of retrieval of basic math facts help students increase working memory capacity and succeed in complex math tasks. Ivon Arroyo is a P.I. or co-PI for NSF and Department of Education research grants that attempt to find principles for the design of digital learning environments for STEM that enhance affective and cognitive outcomes, with an emphasis on girls and students with learning disabilities. She holds a doctorate in Math and Science Education and a Masters and a B.S. in Computer Science.
WestEd PALS STUDY
We are presently recruiting superintendents, principals and teachers for a 4-year efficacy research trial in Massachusetts with 80 teachers (40 treatment/40 control) to test the intelligent math tutoring system, MathSpring.
If you have any questions regarding the study, please contact us.
Thank you for your interest in the Center for Knowledge Communication. Please use the form below to get in touch with us.