Sword Forger is an attempt at taking information gathering in a new direction. Instead of repeatedly browsing through large quantities of formatted data for frequently accessed information, the Sword Forger concept hopes to automate such processes through machine "reading." Simply described, the Sword Forger software, and internally the Sword Forger alogrithms, attempts to search for dynamic text information in a method similar to that of humans readers. This is extremely useful for accessing information that changes over time, but whose basic formatting and context remain the same. An "intelligent" reader can harvest this information with little regard for minor changes in the overall context.

Web pages are an ideal medium for providing dynamic content, and therefore are ideal for my text recognition applications. Our main application, Sword Forger Aggregator, is an implementation of the SF algorithms in which the user can provide some kind of "example text" with which the program searches for similarly patterned text within a given web page. The real value of Sword Forger is that it can be applied with little regard to the location of such textual information within the search content, and therefore makes text searching very flexible.