Developing an Ontology

1
Developing an Ontology
University of Durham University of St Andrews
INSIDE An Institutionally Secure Integrated Data
Environment

• Janet Lavery Cornelia Boldyreff
• Department of Computer Science
• University of Durham
• janet.lavery_at_durham.ac.uk
• cornelia.boldyreff.ac.uk

2
INSIDE

• Collaborative project between the Universities of
  Durham and St Andrews
• Funded by JISC - Joint Information Systems
  Committee
• Building Managed Learning Environments in HE
• Addressing the issues surrounding the development
  and delivery of web based "joined up" system for
  institutions
• Addressing the challenges created by the need to
• incrementally migrate existing services to a
  web-based system
• exploit new technology to provide new value-added
  services
• develop enterprise systems that are both internal
  to the organisations and "joined up" into wider
  inter-institutional networks

3
Progress

• Modelling current institutional Information
  Systems and business practices
• Generic
• Specific
• Identifying areas ripe for improvement
• Based on incremental analysis and implementation
• Designing and implementing value-added services
• Web-based access to current resources and data
• New uses for using existing data

4
Developing a Model

• Working toward a multi-layered model of HE
  Institutions and their business processes for
  engineers to exploit when evolving the existing
  HE systems
• Version 2.0 on web site
• Comprised of three work products
• UML Model in Rational Rose Use Case and Logical
  View
• Thesaurus of generic terms
• Ontology of domain terms (work in progress)
• Business Areas of the Domain
• Undergraduate Registration Process
• Departmental manipulation and use of student
  data.
• XML Student Records
• Value-Added Services

5
Meta-Process Model

• Analyse, Model, Improve, Repeat
• Targeting specific areas of concern in the domain
• Student registration or student data analysis
• Incremental development of value-added services
• Work Products
• evolve as domain knowledge improves
• evolve as domain changes
• Using UML
• Fit for Purpose

6
Meta-Process
7
Initial Analysis
8
Formal Analysis
9
Areas Ripe for Improvement

• Student Oriented
• Module Enrolment
• Electronic Submission
• Assignment Tracking
• Access to Own Record
• Academic Feedback (grades)

• Staff Oriented
• Attendance Details
• Assignment Tracking
• Individual Student Details
• Module Administration
• Role Relevant Information Access
• Resource Planning
• Multiple Student Details
• Cohort Analysis

10
Core Requirements
11
Design and Implementation
12
Hierarchical Relationship Mapping
13
Value-Added Service

• Module Enrolment
• Access via Matriculation Card
• Access via the Web
• Authentication by LDAP
• Based on user login Id
• password required
• managed by IT Services
• Review of Student Personal Data
• Role-Based Information Access

14
Module Enrolment Sample
http//www.dcs.st-and.ac.uk/inside/
15
Value Added Services

• Full Attendance Record
• Lecture, Practical, and Tutorial
• Medical and Good Cause Evidence (absences)
• Diverse Methods of Recording
• Staff manually enter attendance data
• Student review their own attendance on-line
• Marks for Assessed Work
• Identified as provisional marks
• Staff can enter marks
• Students can review marks

16
Web-based access to student info.
17
Value-Added Service

• Electronic Submission - Web Based Assignment
  Tracking
• Staff use is voluntary
• Staff can
• list of assessed work details, hand-in format,
  due dates, extra notes
• collect submitted work
• Students can
• view the list of work to be submitted due dates
• view the list of work they have submitted
• submit their assessed work and work logs

18
Logical Model / Ontology
19
5 Approaches to Ontology Development

• Inspiration - focus is on an individual viewpoint
  of the domain
• Induction - focus is on in-depth knowledge of a
  specific area of a wider domain
• Deduction - focus is on the general principles of
  the domain
• Synthesis - a base set of ontologies are
  identified and used to represent specific
  subsections of the domain and
• Collaboration - the viewpoints of many
  individuals are requested and then represented.
• Holsapple, C. W. and K.D. Joshi, (2002) A
  Collaborative Approach to Ontology Design,
  Communications of the ACM, Vol. 45, No. 2.

20
Meta-Process
21

• Any Questions?
• janet.lavery_at_durham.ac.uk
• www.dcs.st-and.ac.uk/inside