Adaptive Tutoring System Research Articles

Bayesian networks for student model engineering

Bayesian networks are graphical modeling tools that have been proven very powerful in a variety of application contexts. The purpose of this paper is to provide education practitioners with the background and examples needed to understand Bayesian networks and use them to design and implement student models. The student model is the key component of any adaptive tutoring system, as it stores all the information about the student (for example, knowledge, interest, learning styles, etc.) so the tutoring system can use this information to provide personalized instruction. Basic and advanced concepts and techniques are introduced and applied in the context of typical student modeling problems. A repertoire of models of varying complexity is discussed. To illustrate the proposed methodology a Bayesian Student Model for the Simplex algorithm is developed.

Frame based Intelligent Tutoring System with weighted attributes and adaptive hypermedia

In this paper, we present an adaptive tutoring system for students of multiple domains with a web-based interface for flexibility. The displayed media is adapted according to the student's capability and aptitude by evaluating the student according to historical data and quizzes. The media is divided into dynamic frames and the content of each frame is displayed adaptively. A weighted approach is used to evaluate the student's capability, and frames are generated based on a beta distribution. The initial evaluation of the system based on a paper simulation showed encouraging results.

PEGASE, un système tutoriel intelligent générique et adaptatif en environnement virtuel

This work takes place in the framework of the design of a training environment using virtual reality. In this context, we defend the thesis that it is possible to integrate a generic and adaptive intelligent tutoring system (called PEGASE) in a virtual environment, in order to provide pedagogical aid for the learner and pedagogical assistance for the trainer. Our pro- posal is a multi-agent system. It highlights a set of information (actions realized by the learner, domain knowledge...) used by PEGASE to take pedagogical decisions. Our study is focused (1) on knowledge representation for the environment and (2) on an adaptive pedagogical agent suggesting pedagogical assistances.

Teaching College Level Content and Reading Comprehension Skills Simultaneously via an Artificially Intelligent Adaptive Computerized Instructional System

This paper presents a behavioral model for conceptualizing advanced reading comprehension as a “higher order” behavior class. Also discussed are strategies and tactics utilized by an artificially intelligent adaptive tutoring and testing software system designed to shape such comprehension skills while also teaching subject-specific “content” to college students. The system, called MediaMatrix, offers internet delivery of relatively traditional textbook content using highly individualized and adaptive tutorial and assessment procedures (Ray, 1995a; 1995b, 2004). Extant and new research on the effectiveness of this system is presented, with particular emphasis on a preliminary study of students in two small sections of an introductory psychology course. Students were evaluated during early (pre) and late (post) portions of the semester using two equivalent forms of a specially constructed SAT/GRE type reading comprehension test. A statistically significant 17% gain from pre-to-post reading comprehension scores was observed, suggesting that both the behavioral model and the MediaMatrix strategies and tactics for shaping such higher order behaviors merit further research. Practical implications of teaching both specific course content and higher order behaviors such as reading comprehension without direct teacher contact are especially noted.

Modeling Beliefs and Solution Strategies in a Distributed Learning System

We present algorithms guiding the identification of student solution strategies in the Domain-Independent Adaptive Tutoring System (DIATS). The DIATS is a distributed computer-assisted instruction system with a prototype in the domain of psychiatric mental disorders. Problems are solved using differential diagnosis decision trees from the DMS-IV-TR. Student solution strategy identification is performed by the Response Analysis Unit of the modeler. The Domain-Independent Adaptive Tutoring System (DIATS) has a distributed architecture that combines shared-data and client-server styles.

Web-Based Adaptive Tutoring: An Approach Based on Logic Agents and Reasoning about Actions

In this paper we describe an approach to the construction of adaptive tutoring systems, based on techniques from the research area of Reasoning about Actions and Change. This approach leads to the implementation of a prototype system, having a multi-agent architecture, whose kernel is a set of rational agents, programmed in the logic programming language DyLOG. In the prototype that we implemented the reasoning capabilities of the agents are exploited both to dynamically build study plans and to verify the correctness of user-given study plans with respect to the competence that the user wants to acquire.

A fuzzy neural network for item sequencing in personalized cognitive scaffolding with adaptive formative assessment

Adaptive formative assessment is a new approach to implement computerized adaptive learning based on the cognitive scaffolding principle. In adaptive formative assessment, at any stage of a learning session, the system takes into account a student's demonstrated cognitive level to generate the next appropriate formative testing instrument. In this paper, an appropriate strategy to progress students up the cognitive ladder by adaptive item selection is explored using the non-symbolic fuzzy-neural network technology. The proposed model features to learn and memorize good learning paths for different students, and accordingly provide personalized learning sequence for other similar students. The adaptation behavior of the fuzzy neural network to different student categories is investigated by simulated experiments, and its effectiveness is compared with another memory-less binary item selection algorithm. Preliminary results reveal its potential for being an effective adaptive item selection module in adaptive tutoring systems based on cognitive scaffolding with adaptive formative assessment.

적응형 교수 학습을 위한 퍼지 집합 기반 에이젼트 시스템

This paper proposes an agent-based adaptive tutoring system that monitors learning process of learners` and provides learning materials dynamically according to the analyzed learning character. Furthermore, it uses fuzzy concept to evaluate learners` ability and to provide learning materials appropriate to the level of learners`. For this, we design a courseware knowledge structure systematically and then construct a fuzzy level set on the basis of it considering importance of learning targets, difficulty of learning materials and relation degree between learning targets and learning materials. Using agent, monitoring continually the learning process of learners `inferencing to offer proper hints in case of incorrect answer in learning assesment, composing dynamically learning materials according to the learning feature and the evaluation of assesment, our system implements effectively adaptive instruction system. Moreover, appling the fuzzy concept to the system could naturally consider and ideal with various and uncertain items of learning environment thus could offer more flexible and effective instruction-learning methods.

Learning benefits of structural example-based adaptive tutoring systems

This paper illustrates and evaluates a generic adaptive tutoring environment, structural example-based adaptive tutoring system (SEATS), based on the theory of cognitive knowledge acquisition. The system teaches by presenting side-by-side examples and highlighting their common structural components. This technique assists the process of generalization and reduces mapping by surface features, allowing students to apply their newly gained knowledge to different sets of problems. SEATS also implements adaptive presentation based on a straightforward model of student/user performance. SEATS was evaluated with a recursion tutorial used by 117 students in a 1-hour tutorial session. Results indicate that using adaptation in combination with the structural example-based feature produces an effect on rate and extent of learning significantly greater than when the features are used alone, or when both are absent. The study further points out that future evaluations will have to take students' curiosity into account, since many of them are likely to give at least some incorrect answers on purpose, to test the response of the system. The results indicate that building SEATS with simple adaptive mechanisms is an efficient way of teaching electronically.

Structural example-based adaptive tutoring system (poster session) (SEATS)

No abstract available.

Adapting to Learner Attributes: experiments using an adaptive tutoring system

A composite learner model for adaptive tutoring systems has been developed, which combines a model of learner attributes with a simple overlay model of the learner's domain knowledge state. The model of learner characteristics enables several different forms of psychological and background data about the student to be taken into account, and provides a profile of the learner in terms of pedagogically useful attributes. An Adaptive Tutor Using Learner Attributes (ATULA) has been constructed, which uses the composite learner model. The system is able to select for the learner the optimal form of the learning material for the group of topics about to be presented. Experiments carried out with the adaptive tutoring system over 3 successive academic years with two disparate groups of students are described. User records collected during the experiments provide insight into the operation of the model, both at initialisation and during the student's interaction with the system. Previous users’ records also enable comparisons to be made with the result of previous experiments using a non‐adaptive system.

Development of a model of user attributes and its implementation within an adaptive tutoring system

User modelling within tutoring systems often concentrates on the representation of the learner's status with respect to the domain, paying little attention to the user's individual characteristics in terms of capabilities and preferences. A composite learner model, incorporating both domain related data and information about personal attributes is useful in determining not only which items should be presented, but how the student may best be able to learn them. A model of users' individual characteristics has been developed using multivariate statistical techniques as a means of generating user stereotypes from empirical data. Each stereotype has an associated profile in terms of attributes which are useful for the application in which the model is used.

Implementing multiple tutoring strategies in an intelligent tutoring system for music learning

Variation in tutoring strategies plays an important part in intelligent tutoring systems. The potential for providing an adaptive intelligent tutoring system depends on having a range of tutoring strategies to select from. In order to react effectively to the student's needs, an intelligent tutoring system has to be able to choose intelligently among the strategies and determine which strategy is best for an individual student at a particular moment. This paper describes, through the discussion pertaining to the implementation of SONATA, a music theory tutoring system, how an intelligent tutoring system can be developed to support multiple tutoring strategies during the course of interaction. SONATA has been implemented using a hypertext tool, HyperCard II. 1.