Showing papers presented at "Artificial Intelligence in Education in 1997"

Intelligent Tutoring Goes To School in the Big City

Abstract: This paper reports on a large-scale experiment introducing and evaluating intelligent tutoring in an urban High School setting. Critical to the success of this project has been a client-centered design approach that has matched our client's expertise in curricular objectives and classroom teaching with our expertise in artificial inte lligence and cognitive psychology. The Pittsburgh Urban Mathematics Project (PUMP) has produced an algebra curriculum that is centrally focused on mathematical analysis of real world situations and the use of computational tools. We have built an intelligent tutor, called PAT, that su pports this curriculum and has been made a regular part of 9th grade Algebra in 3 Pittsburgh schools. In the 1993-94 school year, we evaluated the effect of the PUMP curriculum and PAT tutor use. On average the 470 students in experimental classes outperformed students in comparison classes by 15% on standardized tests and 100% on tests targeting the PUMP objectives. This study provides further evidence that laboratory tutoring systems can be scaled up and made to work, both technically and pedagogically, in real and unforgiving settings like urban high schools.

Distributed intelligent tutoring on the web

Abstract: The World-Wide Web is an inherently distributed environment. As the web becomes a more important medium for education, we need to consider how different educational systems might interact. Adaptive systems are especially promising in this regard, since they offer the potential to customize themselves to suit students who have not yet used the system. In this paper, we describe our efforts to construct an integrated system which incorporates conceptual instruction through InterBook and problem-solving interaction through PAT Online. InterBook and PAT Online are two separate Web-based adaptive tutoring systems. To achieve a better level of adaptivity these systems interact and exchange information about the user’s progress.

The Competence of Learning Companion Agents

Abstract: One recent approach in developing computer-based learning environments advocates the idea of creating a social context inside the computer. It is claimed that when the learner is engaged into a meaningful dialogue with the software actors his/her learning will benefit. In this paper we concentrate on the collaboration with artificial social actors as peer learners. How "able" should the learning companion agent be in order to maintain the motivation of the human learner to collaborate? It has been argued that "too strong" or "too weak" companion agents may frustrate the human learner to quit the collaboration altogether. This paper describes an empirical study where the learner is able to work with several artificial learning companions - both strong and weak ones. Our empirical data deals with young school children working on elementary mathematics. On the basis of this study we put forward that a group of heterogeneous companion agents at the learner's disposal will increase his/her motivation to collaborate with the agents. This study also suggests that besides the competence of the learning companion agents it is essential to pay special attention to the personal voice of the companion agents in order to keep the human learner interested.

The IRIS Shell: "How to Build ITSs from Pedagogical and Design Requisites"

Abstract: The goal we pursue in our research is to build a shell for helping human instructors to develop intelligent teaching-learning systems in a wide range of domains. We aim to provide a system where a previously defined architecture can be adapted automatically into a new tutor using a set of instructor-generated requirements. Trying to provide a sound basis for this tool, we use a theory of instruction that integrates cognitive processes, instructional events and instructional actions within a three-level framework that relates them. In this paper we extract, through the analysis of both the cognitive theory and the generic architecture, the requirements of the different components of a tutor and integrate them in IRIS (IRakaste-Ikaste Sistema; Teaching-Learning System), a shell for building teaching-learning systems. Moreover, we embed these requirements, cognitive principles, and design requisites in a shell in order that human instructors can follow them easily. Various design issues and an example of building a tutor for mathematical differentiation using IRIS are presented. (http://aied.inf.ed.ac.uk/members/archive/vol_8/arruate/full.html)

Anchored Interactive Learning Environments

Abstract: Advances in computer technology and multi-media systems have led to widespread interest in computer-based instruction and learning environments. The use of video, animation, graphics, and simulation allow the presentation of material in realistic contexts, thus addressing the problems of inert knowledge while promoting constructive and generative learning. But the true potential and benefits of these systems are yet to be realized. Cognitive studies on learning and transfer suggest that concepts acquired in a single context often remain coupled to that context, and are not readily transferred and accessed in other problem solving situations. These findings point to the limitations of traditional, single-context, computer-based instructional environments. This paper introduces an emerging architecture for instruction and learning, Anchored Interactive Learning Environments (AILE). AILE extend well-developed theories of educational practice into the domain of interactive computer assisted instruction. The resulting computer-based environments facilitate sustained learning by active learners. The design and implementation of AdventurePlayer, an Interactive Learning Environment (ILE) anchored in the Rescue at Boone's Meadow (RBM) episode from the Adventures of Jasper series is presented. An experimental study demonstrates the effectiveness of this system in the trip planning domain. (http://aied.inf.ed.ac.uk/members/archive/vol_8/crews/full.html)

Toward computational models of motivation: a much needed foundation for social sciences and education

Abstract: Adamatzky, A. (2001a) Space-time dynamic of normalized doxatons: automata models of pathological collective mentality. A theory of higher level cognition and its relation to visual attention. Wisely non rational-a categorical view of emotional cognitive artificial perceptions. In D. Ca " namero (Ed.), Papers from the 1998 aaai fall symposium: Emotional and intelligent: The tangled knot of cognition (pp. 7-12). Orlando, Florida. ArziGonczarowski, Z. (1999b). Perceive this as that analogies, artificial perception, and category theory. A. (1996b). What are emotions for? Commitments management and regulation within animals/animats encounters.

Supporting Programming by Analogy in the Learning of Functional Programming Languages

Abstract: This paper examines the learning of the functional programming language Standard ML. A common technique used by novices is programming by analogy whereby students refer to similar programs that they have written before or have seen in the course literature and use these programs as a basis to write a new program. We present a novel editor for ML which supports programming by analogy by providing a collection of editing commands that transform old programs into new ones. Each command makes changes to an isolated part of the program. These changes are propagated to the rest of the program using analogical techniques. Many commands are at a level high enough to provide guidance to the novice during program development. We observed a group of novice ML students to determine the most common programming errors in learning ML and restrict our editor such that it is impossible to commit these errors. In this way, students encounter fewer bugs and so their rate of learning increases.

Design, implementation and evaluation of an interface in a Computer Based Learning Environments: the example of PÉPITE

Abstract: The work presented here is part of a multidisciplinary project, called the PEPITE project. Its aim is to set up a system to build student cognitive profiles to evaluate their competences in elementary algebra. We consider here the problems linked to using didactic know-how related to pencil & paper tasks and in particular how to transfer these tasks to a data processing system.