ACM Logo  An ACM Publication  |  CONTRIBUTE  |  FOLLOW    

The Semantic Web and e-Learning

By Chris Daly / April 2009

Print Email
Comments (1) Instapaper
The goal of the Semantic Web is to provide the capacity for computers to understand Web content that exists on systems and servers across the Internet, ultimately adding value to the content and opening rich new data, information, and knowledge frontiers. Billions of Web pages are downloaded daily and are easily understood by humans. The knowledge gap exists for computers as these pages are only structured in the hypertext markup language (HTML) formatting language. When applied to the realm of e-learning, be it content interactions or learner management, the potential to add value is almost limitless.

In essence, the Semantic Web is a collection of standards, data structures, and software that make the online experience more detailed, intelligent, and in some cases, more intense. The components that form the building blocks need a firm foundation to ensure data translated across systems can be recognized. This bedrock comes in the shape of Unicode, which defines one standard for representing all characters through the globe. It also has equivalent codes for future character sets. Once common characters can be identified, resources can be described, conceptualized, and categorized. This is where the Resource Description Language (RDF) and the Web Ontology Language (OWL) come in, providing a language and structure for describing all ideas and concepts in the universe and then relating these to each particular subject area.

In addition to the standards that govern the data and its structure, semantic technologies seek to define the framework and method of communication between systems. The Representational State Transfer (REST) software architecture is already defined for the distributed hypermedia system that is the World Wide Web. Within this structure, the Simple Object Access Protocol (SOAP) provides a standard to allow systems to pass messages between each other while the Universal Discovery Description Integration (UDDI) standard effectively provides a way of defining the index of Web Services available. In addition, the Web Services Description Language (WSDL) describes the functions available to systems communicating via Web servers. Fundamentally, these standards and architecture exist to facilitate the smooth sharing of data across different system and networks.

A third major element, Intelligent Physical Agents (IPAs), is software that is distributed autonomously and exhibits artificial intelligence. This translates as software that can be executed on physically distinct systems, can modify its behavior, and can reason and learn on an on-going basis. This is a key component of the Semantic Web because IPAs will make the intelligent connections between content, mapping relationships, and alerting users and systems to content that previously would not have been identified, or if recognized, would have been discovered accidentally by searching or user recommendation. The Web will essentially be building correlations between defend types of learning interaction regardless of whether the user is online.

E-learning Potential
E-learning is facilitated and supported through the use of information and communications technology, including technology used exclusively in a classroom environment as well as blended learning where technology may be employed in the classroom and then supplemented by external electronic coursework. The third aspect of e-learning-distance learning-allows learners to interact directly online with the digital resource without any face-to-face communication. In distance learning there is often no tutor role and the learner simply works sequentially through content. In all three of these scenarios the learning can be synchronous, where the online interaction between student and tutor takes place in real time, or asynchronous, where the student learning experience is not dependent on real time communication.

The Semantic Web offers learners the possibility of having a wealth of related content delivered to their desktop without explicitly identifying or requesting it. Meaning and associated relationships between content in disparate systems will be continuously evolving. Conceptually related content from learning objects to content stored within Virtual Worlds such as Second Life, would provide a web of complex learning interactions both relevant and interesting to the learner. E-learning facilitators, be they teachers or advisors, can utilize this rich content to enhance the learning experience, allowing them to deliver engaging and relevant courses.

In this age of Web 2.0 where content is user generated and the emphasis is on software components being modular, content including text, audio, and video is generated reviewed, rated, and recommended by users on forums, blogs, and dedicated user groups. While this content does not go through a structured workflow, it offers standalone value as well as adding value to content created by an institution or commercially licensed. This content also provides scope within a semantic context. Technologies are evolving that enable statistical analysis of user-generated content, while tag clouds already exist within many Web 2.0 applications allowing users to add subject keywords to the content they create.

From a commercial perspective publishers are beginning to take on the challenge of implementing RDF which will offer them the ability to easily repackage content to new and emerging markets as topics are related to context. This content can then be delivered to users explicitly as well as making it possible for computer programs like Google who user Web spiders, to search, discover, pick up, collect, analyze, and process information resulting in increased revenues from the viral spreading of structured valuable content. Publishers have been using XML for some years and the Semantic Web will open up new avenues for existing content to be repackaged. Specific e-learning content that uses the Shareable Content Object Reference Model (SCORM), developed as a specific e-learning standard by the U.S. Department of Defense, will further provide a means by which learning objects can be ported seamlessly between different Learning Management Systems (LMS). Integrating learning objects with Web content would deliver new and exciting experiences to the user, as well as content and course creators.

Invigorating Learning
The potential of the Semantic Web could actually revolutionize the learning experience. Roger Schank, who helped found the Learning Center at Carnegie Mellon University, designed a new methodology that eliminates classes, tests, lectures, and even programs themselves. The idea behind the Story-Centered Curriculum (SCC) is to create a story that a student will live in for a certain time period that encompasses everything he or she would do in the given time frame. Schank argues the most effective way to teach new skills is to put learners in the kinds of situations in which they need to use those skills, and to provide mentors who help learners as and when they need it. Effective learners come to understand when, why, and how they should use skills and knowledge. They receive key just-in-time lessons, in such a way that learners will most likely remember the information later when they need it. In a Semantic Web context, learning would be continuously invigorated with the obvious benefits being an increase in the quality of content and the sophistication of student interactions.

The prospect of applying semantic concepts to learning administration as well as direct pedagogy could offer benefits to the institution and the learner. Student and teacher management systems contain human resource and financial information with data about their attendance, timetables, and medical information. Records of students with medical conditions could be kept and treatment options updated when medical authorities authorize new medicines. When the student makes an appointment at the medical center, the school administration system would be informed without any direct communication from the student. Other student information system features such as classroom management and teacher home communication would be updated dynamically. Employment searches could include alerts when a new teaching position becomes available, eliminating the need for database searching or a preconfigured keyword search. Applications for further education could be stored on the institutional system and consequently exchanged directly with the college or university. In a virtual-world setting, students could experience universities and their environments online enabling a more informed choice. Students could also view courses in preferred topics or areas of interest across regional, national, and potentially international boundaries.

This internal and inter-institution data exchange could be facilitated implementing a Service-Oriented Architecture (SOA) approach. It is a relatively new way of joining systems within enterprises, but has gained popularity due to its low cost for integration coupled with flexibility and simplification of setting up. SOA is middleware that allows systems within an organization to expose their application logic. In a semantic scenario all that would be required is to tag the content in RDF and OWL. In layman's terms this means a way for existing applications to speak effectively to each other, understand context and infer relationships, allow HR systems to talk to library systems, or finance systems among many others. A student record system could expose the services for working with enrollment and registration information, which could then be integrated within a VLE or library system. This would not only mean content migrating from one LMS to the other using standards such as SCORM or IMS for learning data, but content understanding and building relationships that live in other trusted system across the institutional network, and by extension other trusted external networks.

There is, however, a major hurdle to a global Semantic Web once there is only one character set, and content is linked to only one Web location by via a Uniform Resource Indicator (URI), and agreements have been reached regarding standards, and a framework facilitating intelligent data exchange has been implemented and universally adopted. The hurdle is security. Additionally commercial license agreements must be adhered to so sophisticated methods of encryption with intelligent systems having the ability to identify secure data as well as the provider and its owner must be prevalent and robust. Lawmakers and government agencies have an influential part to play in promoting security. For these reasons educational organizations should keep data secure while addressing issues around open access, though in principle the way would be clear to integrate systems across intranets and extranets.

Movers and Shakers
We can conclude there is no obvious consensus as to when the Semantic Web will become fully integrated in the e-learning space, but as previously noted it will take a combination of players. Technology companies and organizations will drive the technology that underpins the Semantic Web, supplemented by standards bodies establishing explicit standards in the fields of technology, content, and security. Publishers must provide structured content to enable high-value content, while browsers and applications will take care of content generated by users in Web 2.0 applications or that is freely available and stored using the creative commons licensing.

Educational institutions may well require an internal semantic champion with the requisite technical, business, and e-learning knowledge to broker a successful integration of semantic application with an e-learning strategy. Government agencies and lawmakers need to engender the broad necessity and the vision as well as provide adequate support and development mechanisms for those institutions and innovators wishing to further semantic applications within e-learning. Finally, and perhaps most importantly, the learners and tutors must embrace the new opportunities and pedagogical frontiers that a web of meaning could ultimately deliver.



Comments

  • Thu, 23 Apr 2009
    Post by Doug Wallace

    Have a look at Thinkign Cap. We bill ourselves as the SKM System and everything you say in this article is kind of what we are about.