Selasa, 7 Julai 2009

“You simply cannot beat Google. He is a giant, even bigger than giant!”.

Thats what people say when we talk about Google. But not for Wolfram Alpha. Wolfram Alpha is a website that answers queries and questions based on user inputs. The website was developed by Wolfram Research.
It doesn’t simply return documents that (might) contain the answers, like search engines do. It isn’t just a giant database of knowledge, like encyclopedias. It doesn’t simply parse natural language and then use that to retrieve documents, like Powerset. Instead, Wolfram Alpha actually computes the answers to a wide range of questions — like questions that have factual answers. Wolfram Alpha doesn’t simply contain huge amounts of manually entered pairs of questions and answers, nor does it search for answers in a database of facts. Instead, it understands and then computes answers to certain kinds of questions.

How it works?

Users submit queries and computation question by typing them in a text box. Wolfram Alpha then computes the answers and displays visualizations from a large core knowledge base of structured data.
Wolfram Alpha has been bulit upon Wolfram’s earlier product, Mathematica, which uses computer algebra, symbolic and numerical computation, visualization and statistics. This is run in the bacground of the answer engine so it is able to solve all kinds of mathematical questions providing it in a way that is easy to read. The answer engine is also able to answer natural language fact based questions. For example “How many calories in a Mars Bar?” or more complex questions like “How old was Charles Darwin in 1850?”. It solves this using standardised phrases. For example, “How many calories in a Mars Bar?” is rephrased into “Mars almond bar|amount|one bar|total calories” so the answer engine can display the answer at the top of the page it generates.
The current database that Wolfram Alpha includes thousands of datasets, including all Current and Historical Weather. The datasets had been collected over two years, before the answer engine was opened to the public, and now more information is being added to it, providing an ever expanding number of datasets and knowledge database.

In a nutshell

There is no doubting the complexity and commitment of Wolfram Research, however it needs serious upgrades and modernisations to become the ultimate knowledge database.
For example, if you pit it against Google at the present moment then Google will come away victorious, this is because Wolfram Alpha only provides answers to certain questions, whereas Google provides page after page with information on the subject, albeit the information will quite often be repeated over the numerous pages. Another good example of this are encyclopedias, Wolfram Alpha can be compared to Wikipedia because both display and project their information over the internet. Wolfram Alpha finds the answers to specific questions, however Wikipedia can provide the same results as well as show excellent detailed information and description about the same data subject, as well as background information about links to the main data subject.

SO, this tool will NEVER be able to KNOW EVERTHING. However, it will go mighty close.