Agent-Based Architecture for Intelligence and Collaboration in

Agent-Based Architecture for
Intelligence and Collaboration in
Virtual Learning Environments
Punyanuch Borwarnginn
5 August 2013
Outline
• Virtual Learning Environments
• Problems
• Baseline capturing (Survey results)
• Proposed solution
• Intelligent Learning Environment
• Evaluation
• Plan
Virtual Learning Environments
▫ Web-based learning environments
▫ Support classroom learning
▫ Self-learning
• Examples
▫ Blackboard WebCT
▫ Moodle
Problems
VLEs usually
• lack of assistive feedbacks.
• lack of providing personalisation and adaptivity.
• lack of supporting collaborative tasks.
• itself is not an automated system.
How do we improve VLEs to support these issues?
Baseline Capturing
Survey Results
Purposes
• To understand a students' behaviour and a
classroom style.
• To capture current uses of an online learning
environments that students use in their
learning.
• To evaluate the satisfaction of the current use
in an online learning environments.
• To be able to use these data as orientation data
for establishing issues and requirements of the
project.
Data collection
• Questionnaire
• Faculty of Information and Communication
Technology, Mahidol University, Thailand
▫ 11 Lecturers (44%)
▫ 283 1st-3rd Undergraduates
 valid answers : 277 (40.67%)
What kind of learning could best describe
the students’ learning behavior?
1%
18%
18%
Individual Learning
50%
49%
Collaborative Learning
Mixed
64%
Students
N= 272 ,No answer = 5
Lecturers
Which style of classroom could describe
your class?
9%
4%
18%
Student-centered
45%
51%
Teacher-centered
Mixed
73%
Students
N= 274 ,No answer = 3
Lecturers
Have you experienced any issues or problems
according to this learning behavior and classroom
styles?
90
80
81.82
70
72.73
60
50
59.93
54.78
53.68
%
45.45
40
30
54.55
Students
Lecturers
32.35
27.27
20
10
1.84
0
Student engagement Student performance
(e.g. class/tutorial
attendance, students’
participation, and
etc.)
Student motivation
Learning activities
(e.g. teaching, group
discussion, uses of IT
in learning, etc.)
Other
More details
Lecturers 'views
• Not all students were interested in the activities.
• Students pay no attention.
Students’ views
• The attention span of students is in general quite short
• Sometimes students want to know why we should learn this subject and If
we study well in this subject what can we use benefit (useful) from this.
• Students did not participate with teacher as it should be.
• Lack of resources for teaching.
• Some subjects have many lessons and very board then we want some
activities for making we active to learn.
• Some subjects are difficult to explain in lectures. Other learning
activities could make them easier.
• Students have different backgrounds, profiles, learning styles and
knowledge about their subjects.
Have you ever used learning management systems (LMS),
virtual learning environments (VLE), e-learning systems or
course websites?
• Lectures
▫ 100% Yes
• Students
▫ 89% Yes (247)
▫ 11% No (30)
Frequency of use in different features (Students)
100%
0.41
2.87
4.1
11.89
90%
12.3
1.64
18.33
23.85
27.46
80%
70%
43.03
4.92
25
60%
35.95
41.6
9.02
64.46
23.01
28.33
Never
23.97
50%
36.48
20.92
40%
Seldom
26.89
Occasionally
25.42
9.5
49.59
20%
Always
9.5
30%
13.81
43.03
15.7
18.49
15.29
19.58
28.28
10%
18.41
0%
Download
course
materials
Check course
announcement
Submit
assignment
Take quiz
9.24
6.2
3.78
4.55
Discussion
Forum
Chat room
14.88
8.33
Wiki
Frequently
Questionnaire
Frequency of use in different features (Lecturers)
100%
0
9.09
9.09
9.09
18.18
18.18
90%
80%
54.55
70%
63.64
72.73
60%
27.27
90.91
36.36
50%
Never
100
Seldom
90.91
40%
0
30%
0
18.18
27.27
0
20%
36.36
36.36
27.27
10%
27.27
18.18
0
9.09
0%
Upload course Create course
materials
announcement
Setup
assignments
0
0
0
0
0
Setup quizzes
Discussion
Forum
Chat room
Wiki
Questionnaire
Occasionall
y
Frequently
Features Ranking
Lectures
1. Upload course materials
2. Setup assignment
3. Create course announcement
4. Questionnaire
5. Setup quizzes
6. Discussion Forum
7. Wiki
8. Chat room
Students
1. Download course materials
2. Submit assignment
3. Check course announcement
4. Questionnaire
5. Take quiz
6. Wiki
7. Discussion Forum
8. Chat room
I am satisfied with the current system that I am
using.
80
70
72.73
60
60.42
50
Students
% 40
Lecturers
36.67
30
27.27
20
10
0
0
Agree
2.92
Neutral
Disagree
Suggested Improvements
Lecturers’ views
• Interactive lesson that allows teachers to incorporate
formative assessments into course materials.
• Dynamic web modules for observing student
assignments, performances, and easing up grading
processes.
• More interactive features
• Better User Interface
• Pool of videos (may be imported from Youtube)
categorized by its subjects with a search 'feature'
• Multiple templates of Quiz and Scoring systems
Suggested Improvements
Students’ views
• More functions for supporting collaborative tasks
such as group projects.
• Video Lectures
• More social network integrations
• Search engine
• More contents and activities in Wiki and Forum
Proposed Solution
Hypothesis
Integrating well-designed agent-based systems
can enrich the intelligence responses (adaptivity,
personalisation and task monitoring) during the
learning process in the virtual learning
environment that lead to the better learning
experience.
Proposed Solution
Outcome
Agent-Based Architecture for Intelligence and
Collaboration in Virtual Learning Environments
called “Intelligent Learning Environment” (I-LE)
Intelligent Learning Environment
Objective
• To introduce an agent-based system into a
Virtual Learning Environment
• To personalise and adapt learning materials and
activities based on students’ profiles and
preferences.
• To observe students’ assignments, group
progresses and their performances.
• To assist teachers when it needs their
attention.
Intelligent Learning Environment
Aims
Agents
• Profile Agent
▫ Collect and Update student data
▫ IMS Learner Information Package
• Student Agent
▫ Recommend a student to perform activities
▫ Suggest students to learning resources
• Activity Monitor Agent
▫ Monitor student activities by using state changes
 Student A has created a report B  hasCreated(StudentA, ReportB)
 Report B is reviewed by student C  isReviewed(ReportB, StudentC)
• Teacher Agent
▫ Notify teachers about students progress
▫ Notify teachers when to review and mark assignment
Overview of I-LE
Experimental Design
• Phase I: Baseline capturing
• Phase II: Pre-experiment
• Phase III: Post-experiment
Evaluation
• Deploy in a real learning environment
▫ Comparing their learning experience with the
current virtual learning environment
▫ Undergraduates in Thailand
▫ Interview and survey
Plan
Thank you
Questions
• How to deal with evaluation using a real
environment?
• Are there any suggestions on a student model?