Wolfram Alpha is a new computational knowledge search engine, that promises to derive answers to questions about matters of fact.

Whereas Google and other search engines deliver a disc-full of documents that have the same keyword, Alpha, which launched in May 2009, promises to work out the right answer from facts its owners have put into its database.

Alpha is the brainchild of Stephen Wolfram, a London-born computer scientist who in 1988 developed Mathematica, an all-purpose algorithm-based mathematical modelling tool. He also worked on cellular automata, the neo-Darwinian idea that very simple programs can generate very sophisticated behaviour by following a few simple rules.

The tool is built on Mathematic, which is a computer language that deals with the formal technical concepts for technical computation - as used by R&D labs. Wolfram Alpha uses this to compute human knowledge, according to Wolfram. He said, "The purpose is to see if we can build a system to compute knowledge accumulated by our civilisation."

The underlying data that Wolfram Alpha uses, is picked from expert sources. Mathematica analyses this data automatically and human curators are used as domain experts to check the data.

Wolfram said the computation and algorithms take the fruits of science and engineering and encode them in a mathematical form using Mathematica so that they can be used by Wolfram Alpha.

The third component is natural language process. Finally, Wolfram Alpha uses automated presentation, to enable users to see the results in an accessible way.

Unlike many search engines which get their data from the public domain, that is, from what's on the public web, Wolfram Alpha gets its data from the 'Deep Web'. That's data sources that either require a subscription, or at least some sort of entry point (an account maybe). In simple terms, some are free, and some are 'fee' (guess which is more robust, structured and reliable).

Wolfram Alpha exploits data sources to determine links or relationships between search-terms, and, more importantly, how the likely important terms (might) relate to one another.

Alpha could constitute a giant step forward in how people use computers.

Click here to try it for yourself and see what you think!

Credit: www.computerweekly.com