Computational Knowledge Engine: Wolfram Alpha

Posted on March 17, 2009
Filed Under browser, creativity, online search, online tools, search engine, semantic web | View Comments

knowledge-engine-wolfram-alpha
Photo credit: Wolfram Research

The latest project by Stephen Wolfram is defined as the first “computational knowledge engine”, something capable of answering factual question for you.

The Wolfram engine is described as “a proprietary system based on fields of knowledge, containing terabytes of curated data and millions of lines of algorithms to represent real-world knowledge as we know it”.

The engine is scheduled to sort out of his private demo in May and most details have yet to be released to the public; nonetheless, Nova Spivack, CEO at Radar Networks (which developed Twine) saw the demo and had a chance to talk with Wolfram himself (which is quite rare, given the almost legendary elusive nature of the author)…

What he experienced seems to be something totally different from both Google giant databases or Powerset natural language processing.

Wolfram Alpha is said to compute the answer to questions that have factual answer, like: “How many protons are in a hydrogen atom?,” “What was the average rainfall in Boston last year?,” “What is the 307th digit of Pi?,” or “what would 80/20 vision look like?”…
This is accomplished using built-in models of knowledge, that are actually massive amounts of data about physical laws AND data about the physical world.

Yet users are still provided with a natural language interface for their questions; and the knowledge domain is not about science only, but a much wider range of fields, from art to travel to tech to people.

We’re apparently talking about a completely different approach from Google’s search engine. As Spivack points out, “Google has built a system that can retrieve any document on the Web. Wolfram Alpha is designed to be a system that can answer any factual question in the world.”

And this is done without using semantic web languages such as RDF, OWL and SPARQL – as these kind of ontologies are too hard to handle for such a wide field of knowledge.

Wolfram Alpha is scheduled to be going live in May. Aren’t you looking forward for it?

Sources: Nova Spivack / Twine note, another Twine note

Matt Marchall / Venture Beat

Stephen Wolfram / Wolfram Blog

A New Kind of Science by Stephen Wolfram: pdf online

Related posts from this blog:
Wolfram Alpha Computational Knowledge Engine – First Official Demo
Knowledge Sharing & Collective Intelligence in the Blogosphere
Research Sources for the Deep Web
Discovery Search Engine For Browsers: Juice

Comments

  • I watched the entire demo presentation at Harvard and though I was a bit upset that it didn't show the screen or examples, I still think it's well worth watching and wrote up my thoughts on WolframAlpha and why it's NOT going to be the Google Killer: http://www.sagerock.com/blog/wolframalpha-not-google-killer/

    Thank you for the introduction article and information! It's always exciting to see what's coming up in the search engine world.
  • Thanks you, imagenin
    i agree w/you
    and tnx for the link.
blog comments powered by Disqus