Towards an Automated Adaptive Content Delivery Training System

1
Towards an Automated Adaptive Content Delivery
Training System

• Authored by
• Gregory Vert, Aparna Phadnis,
• Rajasekhar Yakkali Xin Yu
• Department of Computer Science,
• University of Nevada, Reno

2
Overview

• Introduction
• What are Adaptive Learning Systems
• Other Adaptive Learning Systems
• Goal of Our Research
• Fuzzy Logic Fuzzy Systems
• Description of Our Adaptive Learning System
• System Operation
• Architecture of our System
• Fuzzy Neural Networks (FNN)
• Design Structure of FNN
• Stages in the operation of the FNN
• Implementation Details
• Conclusions
• Futurework

3
Introduction

• Learning
• Factors Influencing Learning
• Different Learning Systems
• Limitations of These Learning Systems
• One Solution
• Adaptive Learning Systems

4
What are Adaptive Learning Systems ?

• Consideration for the concept that
• Each Individual is Different
• Each Individual Learns Differently
• Each Individual has Different Needs
• Each Individual has Different Preferences
• Address the fact that Coursework and Training
  should vary accordingly.

5
What are Adaptive Learning Systems ?

• Adaptive learning systems adapt to the users
  automatically based on this concept to make
  learning more effective, increase user efficiency
  and satisfaction

6
Other Adaptive Learning Systems

• Real Adaptive Intelligent Learning System (RAILS)
• Adaptive Learning Intelligent System (ALIS)
• Virtual Adaptive Learning Architecture (VALA)

7
What is Modality?

• Definition
• Types of Modality
• Examples

8
Factors Influencing Learning

• Age
• Gender
• Time of Day
• Personal Preferences
• Content of Learning Material
• Motivation

9
Goal Of Our Research

• To Develop a New Automated Adaptive
• Content Delivery Training System,
• that adapts and changes the modality of the
  presentation of course content,
• based on an individual users learning factors

10
Fuzzy Logic Fuzzy Systems

• A Quick Review
• - Fuzzy Set Theory
• - Membership Functions

11
System Operation

• Operational Stages
• Collection of Data from User
• Building User Profile
• Short Course
• Evaluation / Assessment
• Decision for Modality
• Presentation of Instruction Material
• Evaluation /Assessment
• Repeat Steps 2,3,.. Or Quit Program
• End of Instruction Material

12
Architecture of System
13
Fuzzy Neural Networks (FNN)

• An Intelligent Hybrid System
• Fuzzy Logic Neural Networks FNN
• Why FNN?
• Handles Uncertainties
• Automatically Generates the Rule Base

14
Design Structure of the Fuzzy Neural Network
(FNN)

• Three Main Stages in the operation of the FNN
• Fuzzification
• Inferencing Process
• Defuzzification

15
Stage 1 in the FNN Fuzzification

• Convert Crisp Input to Fuzzy Input using
  Membership Functions

16
Stage 2 in the FNN Inferencing Process

• Using the Rule Base, Process the Inputs to Get
  Fuzzy Output

17
Stage 3 in the FNN Defuzzification

• Convert the Fuzzy Output to Crisp Output

18
Internal Structure of the FNN
19
Implementation Details

• Networked Laboratory with Nodes running,
• Red Hat Linux
• M S Windows XP
• Software Development using,
• M S PowerPoint
• Visual Basic
• C

20
Conclusions

• Based on Fuzzy Logic Neural Networks,
• our Adaptive Learning System,
• Chooses an Appropriate Modality of Presentation
• Offers a Choice to the User to Select Modality
• Presents Instruction Material
• Assesses User Learning
• Improves the Learning Pace of User

21
Futurework

• Integration of Effects of Behavioral
  Cognitive Skills of Users
• Refinement Optimization of the FNN
• Integration of Concepts from Kinesthetics
  Virtual Reality
• Addition of System Scalability Features

22
Thank you