Toward Smart Learning Systems

Posted by Todd Borghesani

For the most part, learning management systems remain, well, dumb. They are database-driven applications that don’t really “help” you learn. While LMSs have enormous value, we do not have learning systems that “think” with you.

If we dare to look into the future, we need to ask, “How should learning management systems evolve?” I think that there are two areas that make the future of learning management systems rather exciting. Central to both of the following ideas is the “Semantic Web.”

The Web is a rich trove of text which can be assembled as structured ontologies — formal descriptions of concepts and relationships. These ontologies help computers mimic human understanding. The idea has been discussed for years, and more than a decade ago Sir Tim Berners-Lee, who invented the underlying software for the World Wide Web, sketched his vision of a “semantic Web.”

The Industry Competency Model is central to designing the LMS interface, user portfolios and training scenarios. “We need to align training with what happens on-the-job.” If you have been in training for any length of time you have heard this over and over. You can only do this if, first, you have a way of modeling someone’s workplace skills. You need an industry competency model do this. When used, we can scale a training scenario’s complexity (all training scenarios interrelated through their shared ontology) and design instruction that can be measured against workforce performance.

Are there areas of study that can inform our approach? Yes. In the interactive entertainment industry, specialized talent systems — character progressions, honor systems, character talent trees, class specific abilities and trainers — can be adapted to the world of professional training.

Learning Management Systems Need a Bit of Intelligence. Give a computer a task that can be crisply defined — win at chess, predict the weather — and the machine bests humans nearly every time. Yet when problems are nuanced or ambiguous, or require combining varied sources of information, computers are no match for human intelligence. Few challenges in computing loom larger than unraveling semantics, understanding the meaning of language.

One reason is that the meaning of words and phrases hinges not only on their context, but also on background knowledge that humans learn over years, day after day.

Is this possible in the near future?

Using cybernetics and semantics, we can build systems that learns as we do. Interacting with the external environment and the learner, the learning system would assess/compare performance data that is recorded via a system of inputs, assess results and formulate the next training scenarios.

Cybernetics is a broad field of study, but the essential goal of cybernetics is to understand and define the functions and processes of systems that have goals and that participate in circular, causal chains that move from action to sensing to comparison with desired goal, and again to action. Studies in cybernetics provide a means for examining the design and function of any system, including social systems such as business management and organizational learning, including for the purpose of making them more efficient and effective.

Imagine a Semantic Learning System. Since the start of the year, a team of researchers at Carnegie Mellon University — supported by grants from the Defense Advanced Research Projects Agency and Google, and tapping into a research supercomputing cluster provided by Yahoo — has been fine-tuning a computer system that is trying to master semantics by learning more like a human.

Its beating hardware heart is a sleek, silver-gray computer — calculating 24 hours a day, seven days a week — that resides in a basement computer center at the university, in Pittsburgh. The computer was primed by the researchers with some basic knowledge in various categories and set loose on the Web with a mission to teach itself.

“For all the advances in computer science, we still don’t have a computer that can learn as humans do, cumulatively, over the long term,” said the team’s leader, Tom M. Mitchell, a computer scientist and chairman of the machine learning department.

The Never-Ending Language Learning System, or NELL, has made an impressive showing so far. NELL scans hundreds of millions of Web pages for text patterns that it uses to learn facts, 390,000 to date, with an estimated accuracy of 87 percent. These facts are grouped into semantic categories — cities, companies, sports teams, actors, universities, plants and 274 others. The category facts are things like “San Francisco is a city” and “sunflower is a plant.”

NELL also learns facts that are relations between members of two categories. For example, Peyton Manning is a football player (category). The Indianapolis Colts is a football team (category). By scanning text patterns, NELL can infer with a high probability that Peyton Manning plays for the Indianapolis Colts — even if it has never read that Mr. Manning plays for the Colts. “Plays for” is a relation, and there are 280 kinds of relations. The number of categories and relations has more than doubled since earlier this year, and will steadily expand.

The learned facts are continuously added to NELL’s growing database, which the researchers call a “knowledge base.” A larger pool of facts, Dr. Mitchell says, will help refine NELL’s learning algorithms so that it finds facts on the Web more accurately and more efficiently over time.

As we blend informal and formal learning, as social media technologies link each of together, we will witness the rise of adaptive learning and collective intelligence. Being “smart” will be redefined.

Related Articles

web 3.0/xWeb (elearnspace.org)
Semantic Technologies for Learning and Teaching in the Web 2.0 era – A survey – Web Science Repository (journal.webscience.org)
Semantic Science (semanticscience.wordpress.com)
7th Extended Semantic Web Conference 2010 – Heraklion – VideoLectures (videolectures.net)

Rate this blog post: