Self-improving mechanism: learning strategy-selection criteria from student model states by concept learning from examples¶

Design · All designs
Evidence · no claims cited

Description¶

The article's central mechanism account is that a tutor can improve itself by inducing selection criteria from its own experience. Criteria "are expressed as characteristics of the student model" and are "elaborated by comparing student model states recorded when a strategy was effective and those recorded when the same strategy was not effective". The comparison is performed by machine learning methods, "more precisely, by learning concepts from examples", so each tutoring episode contributes a labelled example of strategy success or failure.

Design Implications¶

Context¶

Requirements¶

Recorded student model states from episodes where a strategy was effective and where it was not effective
Machine learning methods able to learn concepts from examples

Constraints¶

Criteria are limited to characteristics expressible in the student model

Target Learners¶

geometry learners

Learning Goals¶

adaptive selection of didactic strategies during tutoring

Claims¶

Examples¶

-

Key Sources¶

Dillenbourg, P. (1989). The design of a self-improving tutor: PROTO-TEG. Instructional Science, 18(3), 193–216. https://doi.org/10.1007/BF00053359

Page metadata

Field	Value
`type`	`design`
`id`	`strategy-selection-criteria-from-student-model-states`
`title`	`Self-improving mechanism: learning strategy-selection criteria from student model states by concept learning from examples`
`description`	`The article's central mechanism account is that a tutor can improve itself by inducing selection criteria from its own experience.`
`status`	`review`
`generated`	by `process:settle-candidates` on `2026-10-18`
`trust`	unverified — no one has confirmed this page's content
`sources`	1 — `dillenbourg-1989`

Frontmatter is the page's machine-readable record — see the Schema & Guide.