A natural language-based search tool which claims to
offer an alternative approach to Google for searching information,
is due out in May.
The tool, Wolfram Alpha, is built around a vast repository of
curated data from public and licensed sources. It uses
sophisticated Natural Language Processing algorithms to allow users
to ask questions in natural English.
Internet groups are working on the semantic web to make it
easier for computers to understand the web. British-born physicist
Stephen Wolfram, who developed Wolfram Alpha, has taken a different
approach. Wolfram Alpha uses the algorithms to identify and
catalogue data.
The product was demonstrated to a small audience on Wednesday at
Harvard University's Berkman Center for Internet and Society. Blog
posts on the internet from people who attended the demo have been
positive, but some people are questioning whether the tool may be
too accurate for its own good.
One user on the
Wolfram Alpha blog said, "My guess is that the greatest
advantages of Wolfram Alpha - its precision and accuracy - may
actually cause it to be of limited interest to most people."