Dynamic Contextual eLearning – Dynamic Content Discovery, Capture and Learning Object Generation
from Open Corpus SourcesShay Lawless, Knowledge & Data Engineering Group, Trinity College Dublin
For more information visit: www.cs.tcd.ie/~slawless
Methodology
eLearning across all environments, be they corporate, formal or informal, can be greatly enhanced if the learning experience can be satisfied both on demand and contextually (i.e. tailored to the experience, goals and personal preferences of the individual learner). However, such personalized eLearning systems are typically reliant on proprietary content that has been specifically developed for the system, as it is very difficult to automatically discover, analyse, harvest and reuse open corpus content.
The Problem
Research Goal and ScopeThe goal of this research is to provide dynamic contextual retrieval of open corpus content for eLearning systems. There are many sources of open corpus content available, the primary examples of which are the Worldwide Web and open corpus digital content repositories. The nature of open corpus content is either inherently unstructured or lacks consistency in its structuring. These structural inconsistencies are both semantic, relating to the vocabulary/ontology used, and syntactic, relating to the metadata standards implemented. Metadata mappings to a canonical model, lexical analysis and a fixed vocabulary ontology will be used to ensure consistency and accuracy in content discovery, analysis and description
The aesthetic qualities and presentation of the content retrieved, and IP / DRM are both outside the scope of this research.
It is proposed that a series of content requirements can be extracted from adaptive course generators and course authoring systems. These requirements will be both technical, relating to content structure and format, and semantic, relating to subject and pedagogical needs. Queries can be created from these requirements to source suitable open corpus content. Once suitable content is sourced it will then be harvested using an API. The content can then be manipulated and re-sequenced for ease of use and reuse in the creation of online learning offerings.Initial development on a first prototype of the Learning Object Search and Construct Service has commenced with trialling of lexical analysis tools and semantic annotation tools running in parallel.
Future WorkWork underway or soon to commence
- Initial Prototype of search service
- Trial lexical analysis and automatic annotation software
- Create Metadata canonical model and set vocabulary ontology
The research is undertaken in conjunction with the IBM Centre for Advanced Studies Student Program.
System Architecture RepositorieseLearning Content Repositories can be grouped into two main categories.
•Commercial - IBM Workplace, LydiaLearn, XanEdu, ContentDM
•Open Corpus - Connexions, Maricopa Learning Exchange, Merlot, eduSource Canada, EducaNext
Both categories will be used to source content.
Commercial Repositories
Open Corpus Repositories
WWW
Analysis and Annotation
Lexical Analysis
Metadata Mapping
LO Generation
Query Generation
eLearning System
User Model
User Input
API
Metadata Cache
Harvesting Service
Open Corpus Content Service
NDLR