|
Search tool that matches concepts, not keystrokes, for more
intelligent and useful results.
No
matter how powerful its engine, a database search tool is only as
useful as its ability to understand and fulfill a user's needs. Most
such tools are limited to simple metadata matches. Even tools using
soft queries and fuzzy logic provide more "noise" than added value
-- only compounding the intense frustration of many e-business
customers who expect an online sales tool to be as responsive as a
store clerk.
Using
logic-based reasoning features, Webmind Recommendation allows users
to establish "hard" and "soft" preferences for searches. In addition
to returning the usual exact-match data, Webmind Recommendation can
provide near matches -- not the alternate spellings or random
associations provided by other engines, but intelligent, conceptual
near matches that bring the user results that closely reflect his or
her intent.
For
example, if a user asks for a brown, cotton shirt for under $50,
Webmind Recommendation might also return (apart from exact matches)
slightly more expensive shirts made of cotton blends, in shades that
are close to brown, such as "stone."
Conversely, if a user asks for "kobe beef," references to
Kobe Bryant will not be larded into the results (unless so desired),
because Webmind Recommendation matches concepts, not keystrokes.
Users can also expand or narrow Webmind Recommendation preferences
as they see fit, choosing the precise number of products
returned.
While
Webmind Recommendation's e-business applications are obvious, it is
suitable for use in other enterprise and academic environments in
which conceptual near matches would be useful -- e.g., intranet and
extranet knowledge databases.
Like
all Webmind products, Webmind Recommendation is continuously
optimized (and customized for specific clients and applications) by
our engineers and, via its learning capabilities, from legacy data
and customer interaction. | 
