Ontological Framework for Educational Feedback

1
Ontological Framework for Educational Feedback

• Jelena Jovanovic1, Shilpi Rao2, Dragan
  Gaevic2,3,
• Vladan Devedic1, Marek Hatala2
• 1FON-School of Business Administration,
  University of Belgrade,
• jeljov_at_gmail.com, devedzic_at_fon.bg.ac.yu
• 2School of Interactive Arts and Technology, Simon
  Fraser University,
• srao, mhatala_at_sfu.ca
• 3School of Computing and Information System,
  Athabasca University
• gasevic_at_acm.org

2
Outline

• Introduction
• The problem statement
• The proposed solution
• Learning Object Context Ontologies (LOCO)
• The LOCO-Cite ontology
• An application example LOCO-Analyst
• Mixed Initiative Coding Environment (MICE)
• Extending LOCO-Cite to meet the requirements of
  MICE
• The LOCO-Cite extensions
• Final Remarks

3
The problem statement

• Online educators need appropriate and reliable
  feedback about the students usage of the
  learning materials
• Such a feedback is quintessential for adaptation
  of the learning content
• Learning Management Systems have only limited
  ability for feedback provision they provide
  only
• simple statistics about the tools the students
  have used
• low level interactions with learning content
  (e.g. page views)
• Therefore, teachers lack consequential awareness
  of what their students are working on and how
  satisfied they are with the content

4
The problem statement

• Timely and up-to-the-point feedback is also very
  relevant for learners
• In certain learning domains such a feedback is
  even a condition sine qua non for successful
  learning
• e.g. when learning to program
• Our goal is to provide
• online educators with useful and reliable
  feedback about the students online activities
  and their usage of the learning content
• learners with feedback on how to regulate their
  learning activities

5
The proposed solution

• We claim that
• Semantic Web technologies can improve the current
  state-of-the-art in generating feedback in online
  learning environments
• Ontologies can interrelate information about
  learning objects, learning activities and
  learners captured from various educational
  systems tools
• Semantic annotation (based on domain ontologies)
  as a mean of relating different kinds of learning
  artifacts (e.g., lessons, tests, chat messages)
• Rules and reasoning as means for generating
  feedback out of the available (semantic) data

6
The LOCO framework

• Based on the notion of Learning Object Context
• a specific learning situation, determined by the
  learning activity, the learning content, and the
  learner(s) involved
• aimed at capturing diverse kinds of learning
  situations (i.e., contexts) typically occurring
  in modern e-Learning environments
• Initially aimed at facilitating reusability of
  learning objects and learning designs, later
  extended to also provide support for personalized
  learning
• Integrates different kinds of learning related
  ontologies
• a user model ontology, an ontology of learning
  design, a content structure ontology, domain
  ontologies
• Flexible design (modular architecture) gt easily
  extensible

7
The LOCO-Cite Ontology
8
An Application Example LOCO-Analyst

• Aims at helping teachers rethink the quality of
  the learning content and learning design of the
  courses they teach
• Provides teachers with feedback about
• the learning activities their students performed
  and/or participated in
• the usage of the deployed learning content
• the peculiarities of the interactions among
  members of the online learning community.
• Feedback provision is based on
• Semantic representation of usage tracking data
• Semantic annotation of learning artifacts
• Simple rules for feedback generation

9
An Application Example LOCO-Analyst
Learn more about it from http//iis.fon.bg.ac.yu/
LOCO-Analyst/
iHelp Courses http//ihelp.usask.ca/
10
Extending LOCO-Cite to Meet the Requirements of
MICE

• MICE
• a modeling and reasoning framework aimed at
  assisting student programmers when learning to
  program using an IDE
• monitors and captures students interactions with
  different learning tools in order to provide them
  with relevant and timely feedback
• currently integrates the following systems
  tools
• The BlueJ IDE
• The gStudy Learning Kit
• The iHelp Courses LCMS

11
Extending LOCO-Cite to Meet the Requirements of
MICE

• MICE needs to recognize and respond to the events
  initiated by the learner
• The integrated tools need to track students
  interactions and capture events
• The LOCO-Cite ontology was extended with
  different kinds of events specific for the tools
  that MICE integrates
• Two extensions
• The BlueJ extension
• The gStudy extension

12
The BlueJ extension
13
The gStudy Extension
The LOCO-Cite ontology and the extensions are
available from http//lore.iat.sfu.ca/projects/L
OCO-Analyst/
14
The LOCO-based architecture of MICE
15
Concluding Remarks

• Summary
• The LOCO-Cite ontology defines the notion of
  learning object context by abstracting the most
  relevant concepts from a number of e-Learning
  systems usage tracking data
• LOCO-Analyst demonstrates how data about various
  learning activities can be integrated to provide
  meaningful feedback for teachers
• MICE (with extended LOCO-Cite) captures learners
  activities when studying programming languages to
  assist them in their learning
• Future Work
• Improve the reasoning over the learning context
  data
• Research potentials of the framework for
  improving semantic social networking
• Using Semantic Web Services for on demand
  composition of personalized feedback

16
Ontological Framework for Educational Feedback

• Jelena Jovanovic1, Shilpi Rao2, Dragan
  Gaevic2,3,
• Vladan Devedic1, Marek Hatala2
• 1FON-School of Business Administration,
  University of Belgrade,
• jeljov_at_gmail.com, devedzic_at_fon.bg.ac.yu
• 2School of Interactive Arts and Technology, Simon
  Fraser University,
• srao, mhatala_at_sfu.ca
• 3School of Computing and Information System,
  Athabasca University
• gasevic_at_acm.org