Book · January 994 citations 110 reads 2,264 authors
Download 5.72 Mb. Pdf ko'rish
|
1994 Book DidacticsOfMathematicsAsAScien
Educational goals and system requirements.
1. The global educational goal supported by the tutor is operationalized by an ideal problem class, that is, students are meant to be able to solve all the tasks belonging to this class after tutorial training. 2. The tasks are not one-step tasks of application (of a theorem or a rule), but problem tasks consisting of several steps that are solved by successively applying suitable operators (theorems and rules). 3. There is no deterministic method of solution, that is, there is generally more than one applicable operator for each step in the solution process. Hence, there are, in general, several solution plans or solutions for each task. (This is why tutorial systems for written methods of arithmetics are not among the systems considered here.) 4. The students know which operators are required or permissable for solving the task (transformation rules for transforming terms or equations, geometric theorems for tasks of geometric proof, rules for geometric loci for geometric construction problems). What is to be exercised here is the skill to apply the operators in the context of a problem solution consisting of sev- eral steps. 5. Educational goals are thus: (a) The students should be able to apply the relevant operators of the problem class in the context of a problem contain- ing several steps, (b) The students should know and be able to apply heuris- tic methods to solve problems (e.g., working forward and working back- ward in problems of proof). Global tutorial strategy. 6. The global educational goal is attained by solving problems of the problem class. A growth of learning occurs both through ITS feedback in case of faulty or unsuitable operator applications and through assistance that the students can ask for at any time. It should be noted that task-oriented ITS satisfy the demand formulated by J. R. Anderson that learning should take place within the context of problem-solving (Anderson, Boyle, Farrell, & Reiser, 1984). ITS expert. 7. The ITS expert is a problem solver operating on a knowledge base in which knowledge about the applicability and effect of operators is repre- sented as rule-based knowledge. 8. For each problem of the problem class the expert finds solutions that are appropriate to the knowledge state of the students. 9. The expert is able to check a student solution for correctness and qual- ity. It is able to classify errors as they occur. 10. The expert is "transparent," that is, it uses only knowledge and meth- ods the student is supposed to learn and use (it could not perform Stages 8 and 9 otherwise). It should be noted that subject-matter fields like geometric 219 INTELLIGENT TUTORIAL SYSTEMS proof, geometric constructions, algebraic term transformations, combina- torics, or integral calculus require the ITS to be equipped with a high-per- formance problem solver. The task-oriented ITS ability to provide the stu- dent with an informative error analysis justifies its being called an "intelli- gent" system, and this is at the same time the main difference to nonintelli- gent CAL systems of computer assisted learning (Lewis, Milson, & Anderson, 1987). For J. R. Anderson, the expert in his tutorial systems is the model of an "ideal student" represented by a system of production rules. Real students are represented by deviations from the ideal student, that is, by omitting the rules not yet learnt and/or by adding buggy rules. With this, Anderson in- tends to attain a cognitive student modeling on the basis of his ACT* the- ory. As task-oriented ITS do not pursue the demanding goal of a cognitive student modeling, the costly and inefficient modeling by a production sys- tem can be dispensed with here. Download 5.72 Mb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling