ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we ... more ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we wanted to test the possibility of using machine learning for modeling students’ causal reasoning in a virtual environment designed to assess this skill. Our findings suggest it is possible to use machine learning to emulate student pathways that are able to predict their causal understanding.
We analyze naturally occurring datasets from student use of educational technologies to explore a... more We analyze naturally occurring datasets from student use of educational technologies to explore a long-standing question of the scope of transfer of learning. We contrast a faculty theory of broad transfer with a component theory of more constrained transfer. To test these theories, we develop statistical models of them. These models use latent variables to represent mental functions that are changed while learning to cause a reduction in error rates for new tasks. Strong versions of these models provide a common explanation for the variance in task difficulty and transfer. Weak versions decouple difficulty and transfer explanations by describing task difficulty with parameters for each unique task. We evaluate these models in terms of both their prediction accuracy on held-out data and their power in explaining task difficulty and learning transfer. In comparisons across eight datasets, we find that the component models provide both better predictions and better explanations than t...
The models, methods and tools of quality management in the HYPERTEST distance learning system are... more The models, methods and tools of quality management in the HYPERTEST distance learning system are described. They are based on conception of evident structure of the knowledges and distance learning programs. The making of an external cycle of management is described in detail. It realizes a standard TQM scheme "Plan-Do-Check-Act". The examples of using a HYPERTEST tool's kit to supply a some stages of management cycle are given.
Generalizability of models of student learning is a highly desirable feature. As new students int... more Generalizability of models of student learning is a highly desirable feature. As new students interact with educational systems, highly predictive models, tuned to increasing amounts of data from previous learners, presumably allow such systems to provide a more individualized, optimal learning path, give better feedback, and provide a more effective learning experience. However, any large student/user population will be heterogeneous and likely consist of discernable sub-populations for which specific models of learning may be appropriate. Student subpopulations may differ with respect to cognitive factors, the level and quality of instruction, and many other environmental and noncognitive factors. The era of both “big data” and widely deployed educational software, including Carnegie Learning’s Cognitive Tutor (CLCT) intelligent tutoring system, presents opportunities to analyze increasingly large volumes of data collected during learners’ interactions with educational systems. Th...
ABSTRACT Educational Data Mining researchers use various prediction metrics for model selection. ... more ABSTRACT Educational Data Mining researchers use various prediction metrics for model selection. Often the improvements one model makes over another, while statistically reliable, seem small. The field has been lacking a metric that informs us on how much practical impact a model improvement may have on student learning efficiency and outcomes. We propose a metric that indicates how much wasted practice can be avoided (increasing efficiency) and extra practice would be added (increasing outcomes) by using a more accurate model. We show that learning can be improved by 15-22% when using machine-discovered skill model improvements across four datasets and by 7-11% by adding individual student estimates to Bayesian Knowledge Tracing.
ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we ... more ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we wanted to test the possibility of using machine learning for modeling students’ causal reasoning in a virtual environment designed to assess this skill. Our findings suggest it is possible to use machine learning to emulate student pathways that are able to predict their causal understanding.
Abstract. The integration of adaptive educational systems is changing from an interesting researc... more Abstract. The integration of adaptive educational systems is changing from an interesting research problem into an important practical task. One of the major challenges that need to be accepted on the way is the development of mechanisms for student model integration. In this paper we propose an approach for aligning overlay models of students' knowledge collected by different educational tools relying on different domain representations.
Adaptive link annotation is a popular adaptive navigation support technology. Empirical studies o... more Adaptive link annotation is a popular adaptive navigation support technology. Empirical studies of adaptive annotation in the educational context have demonstrated that it can help students to acquire knowledge faster, improve learning outcome, reduce navigation overhead, and encourage non-sequential navigation. In this paper we present our study of a rather unknown effect of adaptive annotation, its ability to significantly increase student motivation to work with non-mandatory educational content. We explored this effect and confirmed its significance in the context of two different adaptive hypermedia systems. The paper presents and discusses the results of our work.
Abstract The paper analyzes three major problems encountered by our team as we endeavored to turn... more Abstract The paper analyzes three major problems encountered by our team as we endeavored to turn problem solving examples in the domain of programming into highly reusable educational activities, which could be included as first class objects in various educational digital libraries. It also suggests three specific approaches to resolving these problems, and reports on the evaluation of the suggested approaches.
Transfer of learning to new or different contexts has always been a chief concern of education be... more Transfer of learning to new or different contexts has always been a chief concern of education because unlike training for a specific job, education must establish skills without knowing exactly how those skills might be called upon. Research on transfer can be difficult, because it is often superficially unclear why transfer occurs or, more frequently, does not, in a particular paradigm.
ABSTRACT In this paper, researchers report the results of a study of interactive annotated exampl... more ABSTRACT In this paper, researchers report the results of a study of interactive annotated examples (IAEs) in the context of a business programming course. Using code examples is a key element in most course that cover programming, and providing students with asneeded assistance in the form of code annotations is extremely beneficial. In this experiment, a set of nonmandatory examples were made available to students for self-study. Students were able to access the examples at their own pace on their own time.
Achieving transfer–the ability to apply acquired skills in contexts different from those contexts... more Achieving transfer–the ability to apply acquired skills in contexts different from those contexts the skills were mastered in–is, arguably, the sine qua non of education. Capturing transfer of knowledge has been addressed by several user modeling and educational data mining approaches (eg, AFM, PFA, CFA). While similar, these approaches use different underlying structures to model transfer: Q-matrices and T-matrices.
ABSTRACT In this paper we combine a logistic regression student model with an exercise selection ... more ABSTRACT In this paper we combine a logistic regression student model with an exercise selection procedure. As opposed to the body of prior work on strategies for selecting practice opportunities, we are working on an assumption of a finite amount of opportunities to teach the student. Our goal is to prescribe activities that would maximize the amount learned as evaluated by expected post-test success. We evaluate the proposed approach using an existing dataset where data was collected performing random skill selection.
ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we ... more ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we wanted to test the possibility of using machine learning for modeling students’ causal reasoning in a virtual environment designed to assess this skill. Our findings suggest it is possible to use machine learning to emulate student pathways that are able to predict their causal understanding.
We analyze naturally occurring datasets from student use of educational technologies to explore a... more We analyze naturally occurring datasets from student use of educational technologies to explore a long-standing question of the scope of transfer of learning. We contrast a faculty theory of broad transfer with a component theory of more constrained transfer. To test these theories, we develop statistical models of them. These models use latent variables to represent mental functions that are changed while learning to cause a reduction in error rates for new tasks. Strong versions of these models provide a common explanation for the variance in task difficulty and transfer. Weak versions decouple difficulty and transfer explanations by describing task difficulty with parameters for each unique task. We evaluate these models in terms of both their prediction accuracy on held-out data and their power in explaining task difficulty and learning transfer. In comparisons across eight datasets, we find that the component models provide both better predictions and better explanations than t...
The models, methods and tools of quality management in the HYPERTEST distance learning system are... more The models, methods and tools of quality management in the HYPERTEST distance learning system are described. They are based on conception of evident structure of the knowledges and distance learning programs. The making of an external cycle of management is described in detail. It realizes a standard TQM scheme "Plan-Do-Check-Act". The examples of using a HYPERTEST tool's kit to supply a some stages of management cycle are given.
Generalizability of models of student learning is a highly desirable feature. As new students int... more Generalizability of models of student learning is a highly desirable feature. As new students interact with educational systems, highly predictive models, tuned to increasing amounts of data from previous learners, presumably allow such systems to provide a more individualized, optimal learning path, give better feedback, and provide a more effective learning experience. However, any large student/user population will be heterogeneous and likely consist of discernable sub-populations for which specific models of learning may be appropriate. Student subpopulations may differ with respect to cognitive factors, the level and quality of instruction, and many other environmental and noncognitive factors. The era of both “big data” and widely deployed educational software, including Carnegie Learning’s Cognitive Tutor (CLCT) intelligent tutoring system, presents opportunities to analyze increasingly large volumes of data collected during learners’ interactions with educational systems. Th...
ABSTRACT Educational Data Mining researchers use various prediction metrics for model selection. ... more ABSTRACT Educational Data Mining researchers use various prediction metrics for model selection. Often the improvements one model makes over another, while statistically reliable, seem small. The field has been lacking a metric that informs us on how much practical impact a model improvement may have on student learning efficiency and outcomes. We propose a metric that indicates how much wasted practice can be avoided (increasing efficiency) and extra practice would be added (increasing outcomes) by using a more accurate model. We show that learning can be improved by 15-22% when using machine-discovered skill model improvements across four datasets and by 7-11% by adding individual student estimates to Bayesian Knowledge Tracing.
ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we ... more ABSTRACT Causal reasoning is difficult for middle school students to grasp. In this research, we wanted to test the possibility of using machine learning for modeling students’ causal reasoning in a virtual environment designed to assess this skill. Our findings suggest it is possible to use machine learning to emulate student pathways that are able to predict their causal understanding.
Abstract. The integration of adaptive educational systems is changing from an interesting researc... more Abstract. The integration of adaptive educational systems is changing from an interesting research problem into an important practical task. One of the major challenges that need to be accepted on the way is the development of mechanisms for student model integration. In this paper we propose an approach for aligning overlay models of students' knowledge collected by different educational tools relying on different domain representations.
Adaptive link annotation is a popular adaptive navigation support technology. Empirical studies o... more Adaptive link annotation is a popular adaptive navigation support technology. Empirical studies of adaptive annotation in the educational context have demonstrated that it can help students to acquire knowledge faster, improve learning outcome, reduce navigation overhead, and encourage non-sequential navigation. In this paper we present our study of a rather unknown effect of adaptive annotation, its ability to significantly increase student motivation to work with non-mandatory educational content. We explored this effect and confirmed its significance in the context of two different adaptive hypermedia systems. The paper presents and discusses the results of our work.
Abstract The paper analyzes three major problems encountered by our team as we endeavored to turn... more Abstract The paper analyzes three major problems encountered by our team as we endeavored to turn problem solving examples in the domain of programming into highly reusable educational activities, which could be included as first class objects in various educational digital libraries. It also suggests three specific approaches to resolving these problems, and reports on the evaluation of the suggested approaches.
Transfer of learning to new or different contexts has always been a chief concern of education be... more Transfer of learning to new or different contexts has always been a chief concern of education because unlike training for a specific job, education must establish skills without knowing exactly how those skills might be called upon. Research on transfer can be difficult, because it is often superficially unclear why transfer occurs or, more frequently, does not, in a particular paradigm.
ABSTRACT In this paper, researchers report the results of a study of interactive annotated exampl... more ABSTRACT In this paper, researchers report the results of a study of interactive annotated examples (IAEs) in the context of a business programming course. Using code examples is a key element in most course that cover programming, and providing students with asneeded assistance in the form of code annotations is extremely beneficial. In this experiment, a set of nonmandatory examples were made available to students for self-study. Students were able to access the examples at their own pace on their own time.
Achieving transfer–the ability to apply acquired skills in contexts different from those contexts... more Achieving transfer–the ability to apply acquired skills in contexts different from those contexts the skills were mastered in–is, arguably, the sine qua non of education. Capturing transfer of knowledge has been addressed by several user modeling and educational data mining approaches (eg, AFM, PFA, CFA). While similar, these approaches use different underlying structures to model transfer: Q-matrices and T-matrices.
ABSTRACT In this paper we combine a logistic regression student model with an exercise selection ... more ABSTRACT In this paper we combine a logistic regression student model with an exercise selection procedure. As opposed to the body of prior work on strategies for selecting practice opportunities, we are working on an assumption of a finite amount of opportunities to teach the student. Our goal is to prescribe activities that would maximize the amount learned as evaluated by expected post-test success. We evaluate the proposed approach using an existing dataset where data was collected performing random skill selection.
Uploads
Papers by Michael Yudelson