1 Protus 2.0: Ontology-based semantic recommendation in programming tutoring system Presentor: Boban Vesin Boban Vesin, Aleksandra Klašnja-Milićević Higher School of Professional Business Studies Novi Sad, Serbia e-mail: {vesinboban, aklasnja}@yahoo.com Mirjana Ivanović, Zoran Budimac Department for Mathematics and Informatics Faculty of Science, Novi Sad, Serbia e-mail: {mira, zjb}@dmi.uns.ac.rs atija, Croatia, 2012.
28
Embed
Protus 2.0: Ontology-based semantic recommendation in programming tutoring system Presentor: Boban Vesin Boban Vesin, Aleksandra Klašnja-Milićević Higher.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Protus 2.0: Ontology-based semantic recommendation in programming
tutoring system
Presentor: Boban Vesin
Boban Vesin, Aleksandra Klašnja-Milićević Higher School of Professional Business Studies
Novi Sad, Serbiae-mail: {vesinboban, aklasnja}@yahoo.com
Mirjana Ivanović, Zoran Budimac Department for Mathematics and Informatics
Faculty of Science, Novi Sad, Serbiae-mail: {mira, zjb}@dmi.uns.ac.rs
Opatija, Croatia, 2012.
2
Contents
• Introduction
• Personalization of content
• Used technologies
• Protus 2.0 architecture
• Ontologies in Protus 2.0
• Implemented rules
• Learner’s interface
• Conclusion
3
Introduction
• Semantic Web technologies
• Educational environments
• Ontologies
• Ontologies provide a vocabulary of terms whose semantics are formally specified
• Ontologies need additional rules to make further inferences
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
4
Introduction• The major goal of learning systems is to support a
given pedagogical strategy
• Ontologies can be associated with reasoning mechanisms and rules to enforce a given adaptation strategy in learning system
• Protus - PRogramming TUtoring System
• Adaptation of the teaching material and navigation in a course based on the principles of Learning styles recognition for a particular learner
• The main objective of the presentation is to present new version of Protus that completely relis on Semantic web technologies
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
5
Personalization of content
• Customization of content to match characteristics specified by the learner model
• Protus 2.0 provides two general categories of personalization based on recommender systems– Content adaptation – Learner interface adaptation
• Adaptation based on the learning style of the learner
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
6
Learning styles identification
• Index of Learning Styles (ILS)• ILS assesses variations in individual learning
style preferences across four dimensions or domains:– Information Processing: Active and Reflective
learners,– Information Perception: Sensing and Intuitive
learners,– Information Reception: Visual and Verbal learners, – Information Understanding: Sequential and Global
learners.
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
7
Characteristics of learners
Active Reflective
Work in groups Work alone
Preference to try out new material immediately (Ask, discuss, and explain)
Preference to take time to think about a problem
Practical (Experimentalists) Fundamental (Theoreticians) Sensing Intuitive
More patient with details More interested in overviews and a broad knowledge (bored with details)
By standard methods Innovations Senses, facts and experimentation Perception, principles and theories Visual Verbal Preference to perceive materials as pictures, diagrams and flow chart
Preference to perceive materials as text
Global Sequential
Prefer to get the big picture first Prefer to process information sequentially Assimilate and understand information in a linear and incremental step, but lack a grasp of the big picture
Absorb information in unconnected chunks and achieve understanding in large holistic jumps without knowing the details
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
8
Used technologies
• OWL - Ontology Web Language
• Protégé - ontology editor – SWRLTab
• SWRL - Semantic Web Rule Language
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
9
Protus
• Different courses and domains
• Highly modular architecture
• Five central components: – the application module– the adaptation module– the learner model– session monitor– domain module
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
10
Overall architecture of Protus
Session monitor
Application module
Adaptation module Domain module
Learner model
Learner’s interface(interface ontology)
Server side of systemLearner model
ontology
Domain ontology
Task ontology
Teacher’s interface
Teaching strategy ontology
Communication ontology define conection between components
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
11
An excerpt of domain ontology
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
12
An excerpt of resource topology
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
13
Learner model ontology
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
14
Ontology for learner observation
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
15
Teaching Strategy ontology
Class
Personalization
Class
LearningStyle
Class
Condition
Class
AdaptationTypeClass
CurrentGoal
Class
NavigationSequence
Class
Resouce
Class
BehaviourPattern
Class
Decision
Class
Performance
determines
basedOn
generates
generates
Class
Learner
hasLearningStyle
has Performance
isTypeOf
basedOn
basedOn
consistsOf isTypeOf
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion
16
Implemented rules
• In Protus:– the interface elements for sequential
navigation are hidden/shown– Different presentation methods – Adding of links to related or more complex
content
• Three groups of rules:– learner-system interaction rules– off-line rules– recommendation rules
Into Personalisation Technologies Arcitecture Ontologies Rules Interface Conlusion