How to build tutoring systems that are almost as effective as human tutors? Kurt VanLehn School of Computing, Informatics and Decision Systems Engineering Arizona State University 1 Outline Next Types of tutoring systems Step-based tutoring ≈ human tutoring How to build a step-based tutor Increasing their effectiveness Flame 2 Two major design dimensions Personalization of assignments – Non-adaptive – Competency gating » using sequestered assessments » one factor per module – Adaptive task selection » using embedded assessments » one factor per knowledge component Granularity of feedback, hints & other interaction o Assignment (e.g., conventional homework) – Answer (e.g., most regular tutoring systems) – Step (e.g., most Intelligent Tutoring Systems) 3 – Sub-step (e.g., human tutors & some ITS) Example: Pearson’s Personalization Mastering Physics – Non-adaptive Ø Competency gating – Adaptive task selection Granularity Ø Answer – Step – Sub-step 4 Example: Andes Personalization Physics Tutor Ø Non-adaptive – Competency gating – Adaptive task selection Granularity – Answer Ø Step – Sub-step 5 Example: Cordillera Personalization Physics Tutor Ø Non-adaptive – Competency gating – Adaptive task selection Granularity – Answer – Step A step Ø Sub-step 6 Example: Carnegie Personalization Learning’s Tutors – Non-adaptive – Competency gating Ø Adaptive task selection Granularity – Answer Ø Step – Sub-step 7 Carnegie Learning’s skillometer shows knowledge components & current competence Entering a given Identifying units Finding X, any form Writing expression Placing points Changing axis intervals Changing axis bounds 8 Example: Entity-relation Personalization Tutor Ø Non-adaptive – Competency gating – Adaptive task selection Granularity – Answer Ø Step – Sub-step 9 Availability Non-adaptive Competency Adaptive task gating selection Answer-based Thousands Hundreds Few feedback/hints Step-based Hundreds Tens Few feedback/hints (few on market) Sub-step based Tens None None feedback/hints 10
Description: