Transcript TashwinITS
31st October, 2012 CSE-435 Tashwin Kaur Khurana Overview Intelligent Tutoring Systems Components of an ITS Problems Case Based Reasoning in ITS CBR Methods Examples Demo Summary Research areas ITS System that provides personalized tutoring by : Generates problem solutions automatically Represents the learner’s knowledge acquisition processes Diagnoses learner’s approach to the solution Provides advices and feedback Intelligent Tutoring System (ITS) - computer-based training system that incorporate techniques for communicating / transferring knowledge and skills to students. ITS = combination of Computer-Aided Instruction (CAI) and Artificial Intelligence (AI) technology Conventional Model Components of an ITS and their interaction The Student Model The Pedagogical or Tutor Model The Domain Knowledge The Student Model Keeps track of all information related to the learner : ○ ○ ○ ○ Records performance of all the learners Problems assigned Complex uncommon problem solutions Description of approach to the solution with regard to a specific problem Allows system to adapt to learner’s needs Learner’s performance evaluated as a subset of an expert’s performance --- Drawback!! The Tutor Systems Automatic Cognitive analysis Path taken by student Goal Initial competence Learning rate Gets input from the Student model to make its decision to reflect the differing needs of each student. The Domain Knowledge Contains the information the tutor is teaching Concepts Rules Axioms Facts, etc Information on how to link the data for optimum performance of the system Should be updated if there are any changes in the domain ! Individualization Actions required… Problems!!! Problem solving information about each student should be stored for a long time This knowledge must be used for subsequent diagnoses and tutorial decisions! Case Based Reasoning !!!!!!!!!!!!!!! How to represent knowledge so it easily scales up to large domain? How to represent domain knowledge other than facts and procedure (i.e. concepts and mental model)? CBR in ITS Represents the Student model and Domain Knowledge in the form of cases These cases can be used to train the tutorial system for a particular user or someone with similar properties as that user Cases: - Produced by the learner himself Experience from other learners On-demand case generation Predefined cases given by human tutors Case Based ITS- Uses of CBR Problem Solving phase: Find similar problem solved in the past to provide learner with past experience feedback. Case-Based Adaptation: Allows interactive system to adapt to a specific user (i.e CHEF cooking tutor). Can be used to adapt interface component depending on the user’s knowledge of the software Case-Base Teaching: Assists the learner by providing with useful cases for learning new information Types: Static (Pre-defined case base) Adaptive (adapts case base from learner experience) Methods Different type of CBR methods: ○ Classification Approach Systems that provide help on well known pre-analyzed cases ○ Problem Solving Approach Systems that diagnose solution proposed by the learner and to identify the problem solving path used Systems that support planning ○ Planning Approach Systems that support planning Representation of cases: ○ Complete cases= Problem definition + detailed solution ○ Snippets or partial cases= Sub goals + solution within of problems different contexts CBITS: Examples CBITS have been used in many different areas : - Medical: CARE-PARTNER Project Management Math : PAT Jurisprudence Economics Programming : ELM-Art, SQL-Tutor, Chess : CACHET Auto tutor Example 1 :: ELM-ART LISP Tutor Weber and Specht – (1997) Episodic learner model Stores knowledge about the user in terms of a collection of episodes which can be viewed as cases. Every solution stated by the user is diagnosed completely or partially to find problem errors. Keeps track of what components were used and when. ELM-PE and ELM-ART - only systems that use this model ELM Architecture Representation of subject domain Consists of rules and concepts in the form of hierarchically organized frames Concepts: comprise knowledge about: ○ Programming language LISP ○ Common algorithms and problem solving knowledge Consists of: ○ plan transformation leading to semantically equivalent solutions ○ rules Rules: describe different ways to solve the goal stated by the concept Bug rules Example 2 :: AutoTutor Web-based intelligent tutoring system developed by an interdisciplinary research team - Tutoring Research Group (TRG); Student contributions: Text box at the bottom of the screen. AutoTutor response: one or a combination of pedagogically appropriate dialog moves conveyed via synthesized speech, appropriate intonation, facial expressions, and gestures and also text form on the screen. AUTOTUTOR- Authoring Tools Case-based help - a case study replicating the process that teacher would go through to create a curriculum script using the tool. The scenario was created through an analysis of think aloud protocols with actual teachers during the evaluation process. Problems and solutions with the terminology, interface, or concepts were used to generate the case study components, which were then incorporated into an overall composite scenario accessible at any time during the authoring process. Auto tutor •Strengths not purely domain-specific easy creation of curriculum script (no programming skills needed) robust behaviour •Weaknesses shallow understanding only performance largely depends on Curriculum Script Demo!! Elm ART: http://art2.ph-freiburg.de/Lisp-Course Auto tutor: http://rhea.memphis.edu/JSONWebServic e/StartFrame1.htm Auto tutor emotions http://wreg.com/2012/05/01/computertechnology-used-as-tutor/ Summary • ITS “give” personalized instruction • 3 main parts are: • The Student Model The Tutor Model The Domain Knowledge CBITS use different approach: Case-Based Adaptation Case-Based Teaching (Static or Adaptive) Classification Problem-Solving Planning Research Areas Developing Authoring tools Increase modularity of ITS Natural language Modeling Emotion recognition Collaborative Learning ITS are becoming more and more popular as a good assistant to human tutors… 6% of Schools in America are using these tools to teach students in each and every area ! Thank you!!