Improving Intelligent Tutoring Systems: Using Expectation Maximization to Learn Student Skill Levels

Ferguson, Kimberly; Arroyo, Ivon; Mahadevan, Sridhar; Woolf, Beverly Park; Barto, Andy

doi:10.1007/11774303_45

Cited by 17 publications

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pedagogical agents are used for simulating the tutor and student model is used for the simulation of students actions. A simulation system is developed to enable the tutors to pose "what if" questions about the effects of their decisions [8].…”

Section: Related Workmentioning

confidence: 99%

Learning agent based knowledge management in Intelligent Tutoring System

Shanmugapriya¹,

Subhashini²,

Akilandeswari³

2012

2012 International Conference on Computer Communication and Informatics

View full text Add to dashboard Cite

An Intelligent Tutoring System is helpful for the users who want to undertake a learning process. This paper presents an intelligent tutoring system based on learning agent and knowledge agents to improve the learning and teaching process of the tutoring session. The agents involved are knowledge acquirement agent, knowledge sharing agent, knowledge maintenance agent and learning agent. The knowledge acquirement agent is responsible for the construction of the student model and the domain knowledge base. The knowledge sharing agent is responsible for producing personalized teaching web pages to students dynamically. The knowledge maintenance agent is responsible for the maintenance of domain knowledge database and student profile database. The learning agent is used to obtain feedback from the students based on the learning process. The prioritization of the domains is performed depending on the feedback obtained from the students to improvise the tutoring system. This paper aims to provide the faster retrieval of information from the database and the effective teaching learning process to the users. The effectiveness is possible by the frequent updation of the domain knowledge to the prioritized domains.

show abstract

Section: Related Workmentioning

confidence: 99%

Learning agent based knowledge management in Intelligent Tutoring System

Shanmugapriya¹,

Subhashini²,

Akilandeswari³

2012

2012 International Conference on Computer Communication and Informatics

View full text Add to dashboard Cite

show abstract

“…Although some researchers have used machine learning techniques to estimate probabilities for static student modeling systems (Ferguson et al 2006), there have been few attempts to learn probabilities directly from data for more complex dynamic Bayesian models of student performance. Jonsson and colleagues learned probabilities for a DBN to model student performance, but only with simulated data (Jonsson et al 2005).…”

Section: Machine Learning Of Dbnmentioning

confidence: 99%

A multifactor approach to student model evaluation

Yudelson

Medvedeva

Crowley

2008

User Model User-Adap Inter

View full text Add to dashboard Cite

Creating student models for Intelligent Tutoring Systems (ITS) in novel domains is often a difficult task. In this study, we outline a multifactor approach to evaluating models that we developed in order to select an appropriate student model for our medical ITS. The combination of areas under the receiver-operator and precisionrecall curves, with residual analysis, proved to be a useful and valid method for model selection. We improved on Bayesian Knowledge Tracing with models that treat help differently from mistakes, model all attempts, differentiate skill classes, and model forgetting. We discuss both the methodology we used and the insights we derived regarding student modeling in this novel domain.

show abstract

“…For example, the high-school math tutor Wayang Outpost [Arroyo et al 2004] observes help use to form mastery estimates. In this system, machine learning from experimental data helps define model parameters [Ferguson et al 2006].…”

Section: Comparisons To Bayesian Learner Modelsmentioning

confidence: 99%

Tractable POMDP representations for intelligent tutoring systems

Folsom-Kovarik

Sukthankar

Schatz³

2013

ACM Trans. Intell. Syst. Technol.

View full text Add to dashboard Cite

With Partially Observable Markov Decision Processes (POMDPs), Intelligent Tutoring Systems (ITSs) can model individual learners from limited evidence and plan ahead despite uncertainty. However, POMDPs need appropriate representations to become tractable in ITSs that model many learner features, such as mastery of individual skills or the presence of specific misconceptions. This article describes two POMDP representations-state queues and observation chains-that take advantage of ITS task properties and let POMDPs scale to represent over 100 independent learner features. A real-world military training problem is given as one example. A human study (n = 14) provides initial validation for the model construction. Finally, evaluating the experimental representations with simulated students helps predict their impact on ITS performance. The compressed representations can model a wide range of simulated problems with instructional efficacy equal to lossless representations. With improved tractability, POMDP ITSs can accommodate more numerous or more detailed learner states and inputs.

show abstract

Improving Intelligent Tutoring Systems: Using Expectation Maximization to Learn Student Skill Levels

Cited by 17 publications

References 7 publications

Learning agent based knowledge management in Intelligent Tutoring System

Learning agent based knowledge management in Intelligent Tutoring System

A multifactor approach to student model evaluation

Tractable POMDP representations for intelligent tutoring systems

Contact Info

Product

Resources

About