Mobile Topic Maps for eLearning

1
Mobile Topic Maps for e-Learning

• John McDonald Darina Dicheva
• Intelligent Information Systems Group
• Computer Science Department
• Winston-Salem State University, USA

2
Outline

• Mobile Computing in Education
• Ontology-aware e-Learning Systems
• The TM4L-ME Project
• Design
• Topic map nuggets
• Support for topic map nuggets
• System architecture
• TM4L-ME Server (Web Service)
• Mobile client
• Future Work

3
Mobile Computing in Education

• Dramatic growth in the mobile devices use
• Becoming more popular than home computers
• Outnumber personal computers 3 to 1 according to
  the Mobile Technologies and Learning Report
• Mobile web access suffers from interoperability
  and usability problems
• W3C launched its Mobile Web Initiative (W3C MWI)
• m-Learning
• Using devices that allow learners to have a high
  degree of flexibility in location and transport

4
Mobile Computing in Education (contd)

• Difference between e-Learning and m-Learning
• Nature of connection between students and
  teachers
• Device features big differences between the
  output, input, memory and processing power
  requirements
• Mobility of users and devices
• e-Learning always assumes network connectivity
  and sizable bandwidth
• m-Learning must account for limited connectivity
  and bandwidth

5
Topic Maps for E-Learning

• Ontology-aware e-Learning Systems
• Imply shared understanding and consistency
  through the use of domain ontologies
• Ontology-based learning repositories highly
  structured and interoperable
• Topic Maps technology (XTM standard)
• allow semantic categorization of documents and
  the information contained within them
• TM4L (Topic Maps for e-Learning)
• Enables creation, maintenance use of
  ontology-aware courseware conforming to XTM
• Ontologies provide semantic interoperability TM
  ensure syntactic interoperability
• Plug-in architecture, free software

6
The TM4L-ME Project

• Goals
• To create an extensible and reusable mobile
  environment for developing and using educational
  TM
• Remote browsing and editing of educational topic
  maps residing on a server from PDA devices
• Approach
• To extend TM4L to a mobile computing platform
• Preserving TM4L functionality adding, editing,
  deleting topics inserting relationships between
  topics repository browsing
• Web service-based
• Utilizing web services to gain access to TM4L
  functionality

7
The Topic Maps Model

• Knowledge elements classified into objects
  called topics
• Topics are related through associations (have
  types define roles)
• Topics can be addressed by specific resources
  (occurrences)
• Topics, associations occurrences can be scoped
  for different contexts

8
TM4L-ME Design

• Possible solutions
• Converting the full version of the standalone
  TM4L to a J2ME (Java Mobile Edition) app
• Requires an extended focus on mobile device
  memory manipulation
• Server-side TM engine storage and a thin mobile
  client that could only ask the server for the
  information about a particular topic
• Our goal also includes updating a TM from the
  mobile device
• Client-server architecture with a thick mobile
  client that can host a small chunk of the topic
  map we coined the term Topic Map Nugget
  (TmNugget)
• TmNugget assists with handling device limitations

9
TmNuggets

• A TM nugget is defined around a center topic
• Simplest case consists of the center topic only
• Can consists of the center topic along with all
  directly connected to it topics and the
  corresponding associations in the TM
• Typically, TM nuggets are defined to contain not
  only the center topic together with its direct
  neighbours, but also those topics that are
  indirectly connected to it (e.g. with a path of
  length 2, 3, 4, etc.
• The allowed depth depends on the available
  memory of the mobile device
• Definition
• A TmNugget for a specified topic ? and dept n (n
  1, 2, 3, ) is a sub-graph of the topic map
  graph, having a central point ? and eccentricity
  e(?) n.
• A TmNugget is represented in XML format XTM
  extended with an additional element tmNugget

10
TM4L-ME Architecture TmNugget
11
TM4L-ME Architecture
12
TmNugget Support

• Extracting a TmNugget with a specified center
  topic from the topic map
• Involves finding a sub-graph of the TM graph with
  a specified center topic and depth
• Inserting or re-inserting a TmNugget back into TM
• Implemented by applying the standard TM merge
  operation with appropriately set parameters
• Sending a TmNugget from the mobile server to a
  mobile client and from a mobile client to the
  mobile server
• Implemented as a service

13
TmNuggets Communication

• Mobile clients request the TM Server for
  TmNuggets
• Each request contains the device specifications
  and the ID of the requested TM
• TM Server compares its specifications to a known
  list of mobile devices and acknowledges the
  device request
• TM Server creates a unique session ID which
  allows the device to make subsequent requests
  without having to re-identify itself
• TM Server creates the nugget
• TM Server sends the nugget to the requesting
  mobile client

14
Creating TmNuggets

• To create a TmNugget, TM4L Server
• Queries the TM repository to find the requested
  topic map
• Links the located TM to the generated session id
• Uses the requesting device specifications to
  determine the size (depth) of the TmNugget (if
  not specified)
• Extracts the nugget with the specified center
  topic
• Serializes the nugget to XTM format

15
TM4L-ME Architecture
16
TM4L-ME Implementation

• Developing the prototype
• Eclipse Development environment
• Java Micro-Edition 2 (J2ME) platform
• Device Simulators
• Ant Build Tools
• Apache Axis
• Jakarta Tomcat
• Blackberry development toolkit
• Blackberry Simulators
• Build Tools

17
TM4L-ME Implementation (contd)

• TM4L-ME Server (Web Service)
• Java Implementation
• Built using Apache Axis
• Interacts directly with TM4L API
• Handles TmNugget distribution

18
TM4L-ME Implementation (contd)

• Mobile Client
• Targeted the Blackberry platform
• Good user base
• Keyboard, scroll mouse
• Mature development tools

19
Future Work

• Evaluate TM4L-ME in an actual learning
  environment
• Continue adding features of TM4L to TM4L-ME by
  developing new Web Services
• As mobile device capabilities (memory, etc.)
  increase, expand mobile functionality.